borehole geology and sub-surface petrochemistry of the domes … thesis... · 2018-10-15 ·...
TRANSCRIPT
Borehole geology and sub-surface petrochemistry of the Domes area, Olkaria
geothermal field, Kenya, in relation to well OW-922.
Victor Omondi Otieno
Faculty of Earth Science
University of Iceland
2016
Borehole geology and sub-surface petrochemistry of the Domes area, Olkaria geothermal field, Kenya, in relation to well
OW-922.
Victor Omondi Otieno
60 ECTS thesis submitted in partial fulfillment of a Magister Scientiarum degree in Geology
Advisor(s)
Anette K. Mortensen
Björn S. Harðarson
Faculty Representative
Guðmundur H. Guðfinnsson Sæmundur A. Halldórsson
Examiner Dr. Bjarni Gautason
Faculty of Earth Science School of Engineering and Natural Sciences
University of Iceland Reykjavik, May 2016
Borehole geology and sub-surface petrochemistry of the Domes area, Olkaria geothermal
field, Kenya, in relation to well OW-922.
60 ECTS thesis submitted in partial fulfillment of a Magister Scientiarum degree in
Geology
Copyright © 2016 Victor Omondi Otieno
All rights reserved
Faculty of Earth Science
School of Engineering and Natural Sciences
University of Iceland
Sturlugata 7
101, Reykjavik
Iceland
Telephone: 525 4000
Bibliographic information:
Victor Omondi Otieno, 2016,
Borehole geology and sub-surface petrochemistry of the Domes area, Olkaria geothermal
field, Kenya, in relation to well OW-922.
Master’s thesis, Faculty of Earth Science, University of Iceland, pp. 144.
Printing: Háskólaprent, Fálkagata 2, 107 Reykjavík, Iceland, May 2016
Abstract
Olkaria is a high temperature geothermal system located within the central sector of the
Kenya Rift System, and associated with a region of Quaternary volcanism. Recharge into
the system is exclusively by meteoric water, derived primarily from the high-altitude rift
escarpments. Tectonic structures play without a doubt a pivotal role in enhancing fluid flow
within the geothermal system. For this study, data and samples from four wells, OW-905A,
OW-910, OW-917 and OW-922, have been used. Wells OW-905A, OW-910 and OW-917
were drilled within and along the proposed caldera structure. Well OW-922 is a step-out
vertical well, drilled to the east of the proposed caldera structure. This study largely focuses
on well OW-922, while, the other three wells have been included for comparison purposes.
The study presents a comprehensive look at the borehole geology and hydrothermal
mineralisation, sub-surface petrochemistry and microprobe analysis. Lithologies intersected
by the study well include pyroclastics, tuff, rhyolite, basalt, trachydacite, basaltic
trachyandesite, trachyandesite and trachyte. No intrusions are found in this well.
Hydrothermal products in the study well occur both as replacement of primary components
or glassy matrix and as open space filling in veins, fractures and vesicles. Evidence of a
significant cooling process (up to over 110oC) in the geothermal system around the study
well is provided by the difference between measured formation temperature (127oC) and
inferred hydrothermal alteration temperature (epidote 240oC). Cases of calcite overprinting
epidote further points towards a cooling process.
Five feed zones have been deduced from the well and all are categorised as small. Generally,
the well has low permeability and could not sustain discharge after 49 days of heat-up.
Observations from temperature distribution across the field indicates that the study well is
located outside the Olkaria Domes heat source, and consequently, outside the main Olkaria
volcanic zone. Whole-rock chemistry displays a range of compositions of the samples of the
study well, from basalt to trachyte or rhyolite, with the prevalence of the rocks being the
highly evolved derivatives. Major oxides and trace element systematics are dominated by
crystal fractionation, even though other differentiation processes cannot be ruled out.
Evidence of a common differentiation mechanism for the surface and sub-surface rocks
seems likely on the basis of comparison between the two groups of samples. Microprobe
analyses have been carried out on epidote and chlorite in order to assess the chemical
composition and the compositional range of the two minerals. It is found that most of the
epidote are aluminium-rich, whereas, the chlorites are iron-rich varieties. Limited number of
epidote and chlorite has however, impeded possible correlation of the stratigraphy in the four
wells based on major and minor element contents.
Útdráttur
Olkaria er háhitasvæði innan miðhluta gliðnunarbeltis Keníu (Kenya Rift System) og tengist
eldgosavirkni á kvartertíma. Vatn sem streymir inn í jarðhitakerfið er að uppruna regnvatn,
sem fellur aðallega á hálendið meðfram sprungudalnum. Tektóník er vafalaust mikilvægur
þáttur í að auka vatnsstreymi innan jarðhitakerfisins. Gögn og sýni frá fjórum borholum,
OW-905A, OW-910, OW-917 og OW-922, hafa verið notuð í þessari rannsókn. Holur OW-
905A, OW-910 and OW-917 voru boraðar innan og á börmum ætlaðrar öskju. OW-922 er
lóðrétt hola, boruð utan svæðis með fyrri holum, austur af öskjunni. Í þessari rannsókn er
augum aðallega beint að holu OW-922, en hinar holurnar hafðar með til samanburðar.
Ýtarlega er litið á borholujarðfræði, myndun jarðhitasteinda og efnafræði
neðanjarðarmyndana og efnasamsetning steinda var greind með örgreini. Meðal myndana
sem fundust í holunni eru gjóska, túff, rhýólít, basalt, trakýdasít, basaltískt trakýandesít,
trakýandesít og trakýt. Ekki sáust merki um innskot í þessari borholu. Jarðhitaútfellingar
hafa komið í stað upprunalegra steinda og gosglers og sest í opið rými í sprungum, æðum og
blöðrum. Merki sjást um að umtalsverð kæling (um eða yfir 110°C) hafi orðið í
jarðhitakerfinu kringum borholuna út frá muninum á mældu hitastigi (127°C) og áætluðu
ummyndunarhitastigi (epidót, 240°C). Dæmi um að kalsít hafi vaxið yfir epidót benda einnig
til að kæling hafi átt sér stað.
Merki sjást um fimm lektarsvæði í holunni, sem öll flokkast sem minni háttar. Almennt hefur
borholan lága lekt og hélst ekki í blæstri eftir 49 daga hitun. Samanburður á hita í borholum
á svæðinu bendir til að rannsóknarsvæðið sé utan varmagjafa Domes-svæðisins í Olkaria og
þar af leiðandi utan aðal eldgosasvæðis Olkaria. Efnasamsetning bergsýna úr holunni er
breytileg frá basalti til trakýts eða rhýólíts, en þróað berg er ríkjandi. Breytileiki bæði
aðalefna og snefilefna bendir til að magn þeirra stjórnist mest af kristaldiffrun, þó ekki sé
hægt að útiloka önnur þróunarferli. Samanburður á yfirborðs- og borholusýnum bendir til
sams konar þróunarferla fyrir báða hópa sýna. Samsetning epidóts og klóríts var greind með
örgreini til að fá upplýsingar um efnasamsetningu og breytileika í samsetningu þessara
steinda. Epidót reynist álríkt, en klórít er af járnríkri gerð. Þar sem fá sýni með epidóti og
klóríti fundust er hins vegar erfitt að nota breytileika í magni aðalefna og snefilefna í þeim
til að tengja saman jarðlög í borholunum fjórum.
Dedication
This thesis is dedicated to my beloved wife, Vivian and my lovely daughter, Tasha.
vii
Table of Contents
List of Figures ..................................................................................................................... xi
List of Tables ...................................................................................................................... xv
Abbreviations ................................................................................................................... xvii
Acknowledgements ........................................................................................................... xix
1 Introduction ..................................................................................................................... 1 1.1 General overview .................................................................................................... 1 1.2 Overview of well OW-922 ...................................................................................... 3
1.3 Goals of the study .................................................................................................... 4 1.4 Thesis layout............................................................................................................ 5
2 Outline of Geology .......................................................................................................... 7 2.1 Regional geology ..................................................................................................... 7
2.2 Geology of Olkaria Volcanic Complex ................................................................... 9
2.2.1 Surface geology ........................................................................................... 10 2.2.2 Sub-surface geology..................................................................................... 13
2.3 Structural geology and tectonics of OVC.............................................................. 15
2.4 Hydrogeological setting ........................................................................................ 17 2.5 Reservoir properties of Olkaria field ..................................................................... 19
2.6 Summary of drilling history .................................................................................. 20 2.6.1 Phase 1: Drilling of the Surface hole ........................................................... 21 2.6.2 Phase 2: Drilling of the Intermediate hole ................................................... 21
2.6.3 Phase 3: Drilling of Production hole ............................................................ 22 2.6.4 Phase 4: Drilling of Production section (Main hole) ................................... 22
3 Sampling and Analytical Methods .............................................................................. 23 3.1 Sampling ................................................................................................................ 23 3.2 Analytical methods ................................................................................................ 23
3.2.1 Binocular microscope analysis .................................................................... 23 3.2.2 Petrographic microscope analysis ................................................................ 24
3.2.3 X-ray diffraction analysis ............................................................................ 24 3.2.4 Inductively Coupled Plasma-Optical Emission Spectrometry (ICP-
OES) analysis ............................................................................................... 24 3.2.5 Electron microprobe analysis ....................................................................... 25
4 Results ............................................................................................................................ 27 4.1 Lithology of well OW-922 .................................................................................... 27
4.1.1 Pyroclastics .................................................................................................. 27
4.1.2 Tuffs ............................................................................................................. 27 4.1.3 Rhyolite ........................................................................................................ 28 4.1.4 Basalt............................................................................................................ 28
viii
4.1.5 Basaltic trachyandesite ................................................................................ 28 4.1.6 Trachyandesite ............................................................................................. 28 4.1.7 Trachydacite ................................................................................................ 29 4.1.8 Trachyte ....................................................................................................... 29
4.2 Hydrothermal alteration ......................................................................................... 29 4.2.1 Distribution of hydrothermal alteration minerals ........................................ 30 4.2.2 Hydrothermal alteration zonation ................................................................ 39 4.2.3 Distribution of alteration mineral zones in the Domes area ........................ 40
4.3 Feed zones .............................................................................................................. 42
4.4 Alteration and measured formation temperatures .................................................. 46 4.5 Whole-rock chemistry ............................................................................................ 47
4.6 Major and trace element Compositions ................................................................. 47 4.6.1 Major elements ............................................................................................ 48 4.6.2 Trace elements ............................................................................................. 50
4.7 Classification of rock types .................................................................................... 52 4.7.1 TAS classification scheme ........................................................................... 53
4.7.2 Al2O3 vs. FeO classification scheme ........................................................... 54 4.8 Comparison of well OW-922 with surface and other sub-surface samples ........... 59
4.8.1 Comparison with surface samples ............................................................... 59 4.8.2 Comparison with other sub-surface samples ............................................... 60
4.9 Microprobe analysis ............................................................................................... 62
5 Discussion ....................................................................................................................... 71 5.1 Lithological units ................................................................................................... 71 5.2 Hydrothermal alteration ......................................................................................... 71
5.3 Permeability and comparison of temperatures ....................................................... 72 5.4 Rock composition .................................................................................................. 75 5.5 Hydrothermal alteration effects on rock chemistry ................................................ 75
5.5.1 Effects on major element geochemistry ...................................................... 75 5.5.2 Effects on trace element geochemistry ........................................................ 77
5.6 Magmatic differentiation processes ....................................................................... 78 5.6.1 Major elements ............................................................................................ 78 5.6.2 Trace elements ............................................................................................. 79
5.6.3 Assessing other differentiation mechanisms ............................................... 80
5.7 Comparison of surface and sub-surface samples ................................................... 81
5.8 Compositional variation of alteration minerals ...................................................... 82 5.8.1 Epidote ......................................................................................................... 82
5.8.2 Chlorite ........................................................................................................ 84
6 Conclusion and Recommendations .............................................................................. 85 6.1 Conclusion ............................................................................................................. 85 6.2 Recommendations .................................................................................................. 86
References ........................................................................................................................... 87
Appendix A: Lithological description of well OW-922 based on binocular and optical
petrographic analyses. ..................................................................................................... 93
Appendix B: XRD analysis for clay minerals in well OW-922. ........................................ 99
ix
Appendix C: ICP-OES analysis procedure ...................................................................... 103
Appendix D: EMP analysis for chlorite and epidote ........................................................ 105
xi
List of Figures
Figure 1: Map of the KRS showing the location of Olkaria Volcanic Complex and
other surrounding volcanic centres .................................................................... 1
Figure 2: The sub-fields of Olkaria geothermal field............................................................ 2
Figure 3: A section of the Olkaria Domes field showing location of the step-out well
OW-922 and other study wells OW-910, 917 and 905A. Also shown are
other drilled wells. The black triangles delineate the outline of the
proposed ring structure. ..................................................................................... 4
Figure 4: DEM of the EARS. Black lines represent main rift faults, whereas, white
surfaces are lakes ............................................................................................... 7
Figure 5: Map of the EARS showing elevations higher than 1200 m (black areas),
evidencing the main Ethiopian and Kenyan Plateaux (dashed lines). The
black areas mark the Cenozoic rocks of the Rift ................................................ 8
Figure 6: Geological map of the Olkaria Volcanic Complex and the surrounding area
.......................................................................................................................... 10
Figure 7: Surface formations and stratigraphic column of Olkaria Volcanic Complex.
Also shown are a series of eruptive peripheral mafic and trachyte
formations ......................................................................................................... 12
Figure 8: Geographical extents and stratigraphic units of the Olkaria eruptive centres.
The Ol-Njorowa Pantellerite Formation (O1) is not shown ............................. 13
Figure 9: Sub-surface lithotypes of the Olkaria Volcanic Complex ................................... 14
Figure 10: Structural map of the Olkaria Volcanic Complex displaying the orientation
of various fault patterns .................................................................................... 16
Figure 11: Piezometric map showing reservoir recharge areas for Olkaria and other
surrounding geothermal fields along the rift axis ............................................ 18
Figure 12: A vertical schematic sketch of the conceptual model of Olkaria geothermal
system. The suggested heat sources, up-flow zones and temperature cross-
section are shown. The green arrow is directed north ..................................... 19
Figure 13: Drilling progress curve for well OW-922 ......................................................... 21
Figure 14: Lithology, location of feed zones, distribution of alteration minerals and
alteration zones of well OW-922 ...................................................................... 32
xii
Figure 15: X-ray diffractogram of air dried, glycolated and heated sample of smectite.
.......................................................................................................................... 36
Figure 16: Cross polarized image of mixed-layer clay depicting high interference
colours at a depth of 2782 m. ........................................................................... 37
Figure 17: A plane polarized image of fibrous chlorite filling a vesicle at 1826 m. .......... 38
Figure 18: X-ray diffractogram of illite and unstable chlorite showing obscured peak
of 14.18 Å for the three treatments and 7.15 Å peak for air dried and glycol
solvation. The latter peak totally collapses upon heating. ............................... 39
Figure 19: Distribution of alteration mineral zones across a section of the Domes area,
along the traverse A-B. ..................................................................................... 41
Figure 20: Downhole pressure recovery logs for well OW-922. ........................................ 43
Figure 21: Feed zones in well OW-922 as deduced from temperature recovery logs,
drilling parameters and geological observations. ........................................... 45
Figure 22: Comparison of inferred alteration and measured formation temperatures
along with boiling point curve for well OW-922. ............................................. 47
Figure 23: Harker diagrams showing major element variations against SiO2 for
samples from well OW-922 .............................................................................. 49
Figure 24: Variation diagrams for a range of trace elements plotted against the index
of differentiation, SiO2 for samples from well OW-922 ................................... 51
Figure 25: TAS diagram showing the compositional range for sub-surface rocks from
well OW-922. .................................................................................................... 53
Figure 26: Al2O3 vs. FeO classification of sub-surface rocks from well OW-922. ............. 55
Figure 27: TAS classification scheme showing comparison of sub-surface samples
from wells OW-922, 905A, 910 and 917 and, surface samples from Olkaria
and neighboring minor eruption centres (represented by broken lines). ......... 59
Figure 28: A comparison for variation in concentration of selected major element
oxides for wells OW-922, 905A, 910 and 917 .................................................. 60
Figure 29: A comparison for variation in concentration of selected trace elements for
wells OW-922, 905A, 910 and 917 ................................................................... 61
Figure 30: BSE image of epidote from 864 m in well OW-910. Compositional
heterogeneity is clearly seen in this figure with Al-rich sections appearing
darker and Fe-rich sections lighter in color. Numerous compositional
zonation patterns are represented by Al-rich rims and Fe-rich cores. Also,
note the titanite inclusions in epidote. .............................................................. 65
xiii
Figure 31: BSE image of epidote filling a vesicle from 1560 m in well OW-905A.
Epidote is observed occuring as aggregates of small columnar cystals.
Quartz is also noted growing in the vesicle. Black areas in the lower right
hand corner, at the centre and within quartz are epoxy. .................................. 65
Figure 32: A WDS elemental map showing typical compositional zoning of epidote
from 864 m in well OW-910. Grayscale map (a) represent the original
composition, whereas, maps (b), (c) and (d) denote Fe, Sr and Al levels
respectively. ...................................................................................................... 67
Figure 33: A WDS elemental map showing Fe (b) and Al (d) distribution within a
single epidote grain from 2782 m in well OW-922. The darker and lighter
sections in the original compositional map (a) indicates Al and Fe-rich
regions, respectively. Map (c) shows Sr levels. ................................................ 68
Figure 34: BSE image of chlorite replacing plagioclase at 1870 m in well OW-917. ........ 70
Figure 35: A possible conceptual model based on outcome of the study, along the
traverse A-B. ..................................................................................................... 74
Figure 36: Variation diagrams showing the concentrations of Na2O, Al2O3, K2O and
MnO against SiO2. ............................................................................................ 76
Figure 37: Variation diagrams showing the concentrations of Rb, Sr, La, Ni, Ba and
Y against Zr. ..................................................................................................... 77
Figure 38: A plot of La vs. Zr. and Y vs. Zr. showing different magmatic differentiation
processes. The dotted and continuous lines indicates fractional
crystallisation and other petrogenetic processes, respectively. ....................... 80
Figure 39: Harker's variation diagrams between various pairs of incompatible trace
elements displaying fractional crystallisation (dotted line). The flexure
from the main trend (continuous line) represent other magma
differentiation processes. .................................................................................. 82
xiv
xv
List of Tables
Table 1: Utilization of geothermal energy for electric power generation in Olkaria. .......... 3
Table 2: Drilling phases and casing design for well OW-922. ........................................... 20
Table 3: Alteration of primary minerals and order of replacement in well OW-922 ......... 30
Table 4: Interpreted permeable zones based on geological observations. ......................... 46
Table 5: Major and trace element results for well OW-922 based on ICP-OES analysis.
.......................................................................................................................... 56
Table 6: Representative EMP analyses of epidote from wells OW-922, 910 and 905A. .... 64
Table 7: Range and average molar proportion of pistachite (Xps) component in
epidote............................................................................................................... 66
Table 8: Representative EMP analyses of chlorite for wells OW-922, 910, 917 and
905A. ................................................................................................................. 69
Table 9: Molar ratio of chlorite from wells OW-922 and 910. ........................................... 70
xvii
Abbreviations
AUs- Additional Units
BP- Before Present
BSE image- Backscattered Electron image
CKPP- Central Kenya Peralkaline Province
CT- Cellar Top
DEM- Digital Elevation Model
EARS- East African Rift System
EMP- Electron Microprobe analysis
GWDC- Greatwall Drilling Company Ltd.
HFSE- High Field Strength Elements
ICP-OES- Inductively Coupled Plasma-Optical Emission Spectrometry
ITEs- Incompatible Trace Elements
ÍSOR - Iceland Geosurvey
ka- Thousand years
KenGen- Kenya Electricity Generating Company Ltd.
KRS- Kenyan Rift Sytem
LILE- Large Ion Lithophile Elements
LREE- Light Rare Earth Elements
Ma- Million years
Masl- Metres above sea level
MER- Main Ethiopian Rift
MLC- Mixed Layer Clay
MWe – Mega Watt electrical
OD- Outside Diameter
OVC- Olkaria Volcanic Complex
OW- Olkaria Well
POOH- Pulling Out of Hole
ppm- parts per million
RIH- Running In Hole
RKB- Rotary Kelly Bottom
ROP- Rate of Penetration
TAS- Total Alkali versus Silica
TD- Total Depth
WOC- Wait on Cement
WDS- Wavelength Dispersive Spectrometer
Xps- Mole fraction of pistachite
XRD- X-ray diffraction
xix
Acknowledgements
My heart felt gratitude goes to my employer, Kenya Electricity Generating Company
Company Ltd. (KenGen) for nominating and providing me with the study leave to pursue
further studies at the University of Iceland. Sincere thanks goes to the Icelandic Government,
the United Nations University- Geothermal Training Programme (UNU-GTP) through its
able Director, Mr. Lúdvik S. Georgesson for funding this study. To the entire warm staff of
UNU-GTP, Ingimar Haraldsson, Thórhildur Ísberg, Málfríður Ómarsdóttir, Markús Wilde,
and Rósa Jónsdóttir, thank you for your technical support and for making me feel always at
home particularly during the harsh winter periods. Rósa you played a key role in delivering
the required research papers and books, even from outside the UNU-GTP library fairly fast.
I am deeply obliged to you and will always be.
I owe a large debt of gratitude to my supervisors; Guðmundur H. Guðfinnsson, Sæmundur
A. Halldórsson, Anette K. Mortensen and Björn S. Harðarson for the many insightful
discussions, constructive comments and guidance during my study period. Your countless
words of support utterly encouraged and motivated me. You helped me shape the line of my
thought with very many intellectual suggestions. It was a great pleasure and honour to work
with you and indeed I am very thankful. In the same token, I thank Dr. Hjalti Franzson and
Dr. Kristjan Saemundsson for the innumerable discussions we held together regarding the
various aspects of this study. Warm thanks are also extended to Atli Hjartarson for
preparation of the microprobe samples.
To my fellow UNU MSc. and Ph.D. students who form a long list, I am grateful to you all
for the times we spent encouraging each other and sharing common problems. In particular,
your ingenuity and hardwork generated the insights that taught me how the World of science
works. I recognize the unstinting support extended by my colleagues at KenGen in terms of
timely data delivery. Joyce Okoo, Xavier Musonye, Danson Warui, Michael Mwania, David
Wanjohi, Edwin Wafula and Lawrence Onyango, just to mention a few, you are indeed
awesome people and I am greatly indebted.
To my family, thank you for your encouragement, infinite patience during my long absence
and for allowing me a peaceful work time. Your love inspired me every day. Above all,
Glory be to the Almighty God. None of this work would have been possible without his
guidance and protection.
1
1 Introduction
1.1 General overview
Olkaria geothermal field is a high-enthalpy geothermal field (temperature > 200o C at 1000
m depth), associated with a region of Quaternary volcanism in the Kenyan Rift System
(KRS). The Olkaria geothermal field has been extensively explored and developed since the
1980s. Over the past 35 years, the geothermal field has been the study site for new and
innovative geological and geothermal research in Kenya. The geothermal system is located
in the Olkaria Volcanic Complex (OVC), a relatively young (~ 20 ka), salic volcanic
complex, characterised by the eruption of peralkaline rhyolites (Omenda, 1998; Macdonald
et al., 2008). Nine other similar
Quaternary volcanic centres
occur in the axial region of the
KRS and are potential
geothermal resources according
to surface exploration studies
(Simiyu, 2010). The OVC is
located in the central sector of
the KRS, ~ 125 km northwest of
Nairobi city and ~ 40 km
southwest of Lake Naivasha (0°
53’S; 36° 18’E) (Figure 1). The
extent of the OVC is estimated
to be ~240 km2 (Clarke et al.,
1990). It is bound in close
geographical proximity to the
north by Eburru, a
predominantly pantelleritic
volcanic complex, to the east
and south by Longonot and
Suswa volcanoes, respectively.
However, chemical features
indicate that all these four
adjacent volcanic complexes,
situated less than 50 km apart are
quite distinct and derived from
separate magmatic systems (e.g.
Macdonald and Scaillet, 2006).
Unlike Eburru, Longonot and
Suswa, which are complexes
with a distinct caldera, the OVC
lacks a proven caldera structure.
Figure 1: Map of the KRS showing the location of Olkaria
Volcanic Complex and other surrounding volcanic
centres (modified from Ofwona et al., 2006).
2
Various scientific studies have been conducted to establish the existence of the caldera
structure with little success. Some scholars (e.g. Simiyu and Keller, 2000) have postulated
that the group of coalesced rhyolitic domes in the eastern and southern segments of the
volcanic complex mark the rim of a caldera structure.
Other researchers (e.g. Bliss, 1979) have argued that the great thickness of comenditic lavas
(at least 1 km from borehole data) may well suggest the existence of a central comendite
volcano in Quaternary times. Such a view would be consistent with the voluminous
outpourings of comenditic lavas observed on the surface within the OVC. While this
interpretation seems possible, the situation is complicated by the lack of ignimbrite deposits
within and in the surrounding area of OVC.
Exploitation of geothermal resources at Olkaria began in the early 1980s and continues to
date with more than 200 wells having been drilled, with a depth range of between 950 and
3200 m. The Olkaria geothermal field has been divided into seven sub-fields with Olkaria
hill (Figure 2), a prominent geological feature about 340 m high and a basal diameter of 2
km being the reference point. Each of these sub-fields has been explored either partially or
completely. They include Olkaria east, Olkaria northeast, Olkaria northwest, Olkaria
southwest, Olkaria central, Olkaria southeast and Olkaria domes (Figure 2).
Olkaria east sub-field was the first part of the system to be developed in 1981. Exploitation
of the extensive geothermal resource at Olkaria is mainly for electricity production at four
conventional power plants; the Olkaria I, Olkaria I (AUs 4 & 5), Olkaria II, Olkaria III
(OrPower) and Olkaria IV power stations. In addition, there are several wellhead generators
installed and are operational within the geothermal field. As of the end of February, 2016,
Figure 2: The sub-fields of Olkaria geothermal field (modified from Otieno and
Kubai, 2013).
3
the aggregate installed generating capacity of the combined Olkaria stations is estimated at
~ 654 MWe. Resource consent for a 140 MWe power plant development at Olkaria V was
granted by a Board of Inquiry process in May, 2015, with inauguration of construction work
slated to begin in the course of this year. Although other opportunities for direct use are
being explored, the activity is currently largely restricted to greenhouse heating (i.e. Oserian
Development Company) and balneology (Olkaria Spa). A summary of the power plants,
including the year of commissioning and the effective installed capacity, is listed in Table 1.
Table 1: Utilization of geothermal energy for electric power generation in Olkaria.
Power Plant
Name
Year
Commissioned
No. of
Units
Status Total Installed
Capacity
Total Running
Capacity
MWe Mwe
Olkaria I 1981 (15 MWe)
1982 (15 MWe)
1985 (15 MWe)
3 Operating 45 45
Olkaria II 2003 (70 MWe)
2010 (35 MWe)
3 Operating 105 105
Olkaria I (AUs
4 & 5)
2015 2 Operating 150 140
Olkaria III
(OrPower)
2000 (53 MWe)
2008 (39 MWe)
2014 (17 MWe)
2016 (30 MWe)
- Operating 150 140
Olkaria IV 2014 2 Operating 140 140
Olkaria
wellheads
2012- 2015 - Operating ~ 64 ~ 64
1.2 Overview of well OW-922
This study largely focuses on well OW-922, in addition to wells OW-905A, 910 and 917.
Well OW-922 is a step-out vertical well drilled between April and September, 2014 at
surface coordinates X= 208675; Y= 9899449 and Z= 2126 masl. It is located approximately
2.5 km directly east of the proposed Olkaria ring structure (Figure 3). The purpose of drilling
this well was to appraise the resource potential of the mentioned area so as to expand
production drilling to the plain outside the ring structure. This has been necessitated by the
renewed focus by the Kenyan government on development and utilisation of the geothermal
resources for electricity generation. The government has identified additional supply of
renewable energy as a fundamental foundation of achieving its Vision 2030 blue print. The
latter seeks to transform the country into a newly industrialised, middle-income economy.
As a result, Kenya Electricity Generating Company Ltd. (KenGen), the leading power
producer in the country, has embarked on efforts to escalate development of geothermal
resources in sync with its geothermal expansion programme. On the other hand, wells OW-
905A, 910 and 917 were drilled within and along the proposed Olkaria caldera structure
between November, 2007 and December, 2012. These wells have been adopted for this study
based on various aspects; specifically owing to the readily available data on borehole
geology and sub-surface petrochemistry from a previous study (Musonye, 2015). Until
4
recently, there were no prior sub-surface petrochemical data (major and trace elements) since
past petrochemical studies were confined only to surface rocks (e.g. Omenda, 1997;
Macdonald et al., 2008; Marshall et al., 2009).
1.3 Goals of the study
The specific aims of this study are three fold;
1. To study the borehole geology of well OW-922 by analysis of drill cuttings in order to
identify the stratigraphy and the sub-surface geological formations. The borehole
geology study will clarify the geothermal conditions in the system east of the Domes
and lead to comprehensive understanding of the hydrothermal mineralisation,
temperature conditions based on alteration minerals and temperature loggings, major
sources of permeability and water-rock interaction processes.
2. To investigate the chemical characteristics of the rocks in well OW-922 and attempt to
decipher the processes of their petrogenesis. These will be compared with petrochemical
data from wells OW-905A, 910, 917 and surface samples from Olkaria and neighboring
minor eruption centres to establish any petrogenetic connection.
Figure 3: A section of the Olkaria Domes field showing location of the step-out
well OW-922 and other study wells OW-910, 917 and 905A. Also shown are other
drilled wells. The black triangles delineate the outline of the proposed ring
structure.
5
3. To assess the chemical composition and variation in compositional range of epidote and
chlorite, as well as the variables likely influencing the compositional changes.
1.4 Thesis layout
Chapter 1: In this chapter, the project area is introduced, including the sub-divisions, total
installed capacity and location of the study wells. The objectives of this study and the outline
of the thesis are presented.
Chapter 2: Provides an overview of the regional geology of the East African Rift System.
An elaborate description of both surface and sub-surface geology, structural and tectonic
setting, hydrogeological setting and reservoir characteristics of the study area. Also
presented is the drilling history of well OW-922.
Chapter 3: Provides an overview of the analytical methods used to achieve the project
objectives. Analytical techniques spanning from binocular microscope to electron
microprobe analysis are described.
Chapter 4: Presents synthesis of borehole geology, whole-rock chemistry and electron
microprobe analytical results. In borehole geology, the lithologic units intercepted by the
drill hole are discussed. Also discussed are the distribution of hydrothermal mineralisation,
feed zones and a correlation between inferred alteration and measured formation
temperatures. In whole-rock analysis, a petrochemical classification of the sub-surface rocks
in the study well and the compositions of major and trace elements are explained in detail.
Chlorite and epidote chemical compositions derived by electron microprobe (EMP) analysis
are detailed.
Chapter 5: Provides an in-depth discussion based on the outcome of the analytical results.
Chapter 6: This chapter incorporates the conclusions based on the outcome of the findings
envisaged in all the analyses and future frontiers recommended.
7
2 Outline of Geology
2.1 Regional geology
The East African Rift System (EARS) is considered an archetypal continental rift in the
initial stage of continental breakup (Achauer and Masson, 2002) that is still in close
proximity to the underlying thermal anomaly (mantle plumes). The evolution of the EARS
is strongly believed to be structurally controlled (Smith and Mosley, 1993) by the
exploitation of weak
collisional zones at the contact
between the Proterozoic
orogenic belts and the Archean
Tanzanian Craton by the rift
faults (Shackleton, 1996).
South of Turkana (Figure 1),
the rift is divided into two
distinct branches that encircle
the robust Tanzanian Craton:
an older, volcanically active
eastern branch and a younger,
much less volcanic active
western branch (Ring, 2014)
(Figure 4). The eastern branch
extends from the Red Sea in
the Afar region, straddles
through the Main Ethiopian
Rift (MER), the Kenyan
(Gregory) Rift and the basins
of the Tanzania divergence
before terminating further
south in Beira, Mozambique, a
total distance of more than
4000 km. By contrast, the
western branch extends over a
distance of more than 2100 km
from Lake Albert (Mobutu) in
the north to Lake Malawi
(Nyasa) in the south
(Chorowicz, 2005).
The zone is typically trough-like, ~ 40 to 65 km wide, and traverses two broad, elongated
domal uplifts in Ethiopia and Kenya. The elevation of the rift floor is highest in the central
part of the domes, 2000 m in Kenya and 1700 m in Ethiopia, and decreases progressively to
the fringes. The onset of topographic uplift in the EARS is poorly dated. However, it
certainly preceded graben development (Baker et al., 1972). The evolution of EARS dates
Figure 4: DEM of the EARS. Black lines represent main
rift faults, whereas, white surfaces are lakes (modified
from Chorowicz, 2005).
8
back 30-45 Ma (Ring, 2014). Rifting is generally assumed to have developed as a result of
the upwelling of a voluminous mantle plume (the African Superswell), which continuously
impinged upon the base of the stretched continental lithosphere (Ebinger and Sleep, 1998).
The driving force for the process has been ascribed to a hot, relatively fertile asthenospheric
plume or diapir emanating from the top of the African Superswell in the upper few hundred
kilometres of the mantle (Ebinger and Sleep, 1998). To date, considerable uncertainty still
exists about the number of plumes influencing magmatism and tectonics beneath the EARS.
Specifically, the depth extent and continuity of the hot asthenospheric material remains a
tantalizing question which has evoked much speculation and debate (Chorowicz, 2005).
Based on plate analysis, the rift marks a line of very slow crustal spreading, extension being
more active in the Red Sea – Gulf of Aden area, approximated at ~ 2 cm/year. In the Main
Ethiopian rift it is estimated at ~ 1
mm/year, and further less than 1
mm/year in the Kenyan Rift and
southwards (Ring, 2014).
Propagation of the rift began as a
chain of marginally warped
depressions, which were
accentuated as domal uplift
proceeded, until, in mid-Miocene
to early Pliocene times, faulting
produced asymmetrical grabens.
The final uplift phase in the early
Pleistocene was accompanied by
major graben faulting, and
subsequent faulting has intensely
fractured the floor of the rift along
an axial zone marked by caldera
volcanoes. Seismic velocity
information from refraction
measurements and tomography
experiments, reinforced by
gravity studies (Ebinger, 2005),
indicate that the eastern rift lies
along a zone of progressive
crustal thinning with local crustal
disruption.
According to Baker et al. (1972) the uplift of the Ethiopian and Kenyan domes (Figure 5)
has been synchronous in three major pulses of late Eocene (44-38 Ma), mid-Miocene (16-
11 Ma), and Plio-Pleistocene (5-0 Ma) age. Volcanism of intermediate and salic type
displays some relation to uplift in time and space and to the onset of graben faulting.
However, major flood basalt extrusions in the early Tertiary in Ethiopia were related to
massive crustal warping along the future rift margins (Williams, 1970). The volcanism
associated with EARS is overwhelmingly alkaline, and at some volcanoes a strongly alkaline
fractionation series is distinguishable from a more mildly alkaline series (Williams, 1970).
The flood phonolites, trachytes, rhyolites, and ignimbrites of Kenya, and the pantelleritic
Figure 5: Map of the EARS showing elevations higher
than 1200 m (black areas), evidencing the main
Ethiopian and Kenyan Plateaux (dashed lines). The
black areas mark the Cenozoic rocks of the Rift
(Macdonald et al., 2001).
9
ignimbrites of Ethiopia, could have resulted from anatexis of a mantle-derived accreted layer
at the base of the crust (Baker and Wohlenberg, 1971).
The evolution of the KRS, a complex graben ~ 40-65 km wide bounded by major rift faults
arranged en echelon, dates back to the East African orogeny. The rift bisects the Kenya
domal uplift, which itself is superimposed on the eastern margin of the East African plateau
(Baker et al., 1972). The KRS is also located close to the boundary of the Tanzanian Craton
and the Pan-African Mozambique shear belt. Volcanism associated with the KRS started
during the Miocene epoch (~ 24 Ma), resulting in widespread basaltic eruptions in the
Turkana topographic depression, a failed Mesozoic rift system (Nyblade and Robinson,
1994). The Miocene basalts were subsequently faulted and followed by the relatively
extensive, reliably dated flood phonolites. The latter are of chemically homogeneous
composition and were erupted from the crest of the Kenya dome in the late Miocene and
early Pliocene (Baker et al., 1972). The total volume of eruptive rocks is estimated to be >
220,000 km3, with a thickness of up to 900 m, as revealed by faulting at the rift margins. In
an important contribution, Ebinger and Sleep (1998) suggest the flood phonolites to be a
product of partial melting of the lowermost part of the crust.
Volcanism off the rift axis was concurrent with the rifting process and is responsible for the
formation of the vast, uplifted, off-rift volcanic centres, namely Mt. Kenya, Chyulu and Huri
hills, located on the eastern flanks (Figure 1). It has been observed that these three sites of
domal uplift are not rifted, and, support a model whereby doming predated rifting (Wilson,
1989). Further block-faulting of the Miocene volcanics produced noticeably massive and
extensive, more evolved eruptions of trachytic ignimbrites in the central area, provisionally
assigned to lower Pliocene (Ring, 2014). The trachytic ignimbrites formed the Mau and
Kinangop tuffs. A third faulting episode, which followed the ignimbrite eruptions, resulted
in the formation of the graben structure, as it is known today (Baker et al., 1972). In the
developing graben, successive fissure eruptions produced approximately 900 m thick layers
of trachytes, basalts, basaltic trachyandesites and trachyandesites. The plateau rocks that
filled the developing graben were subsequently block-faulted to create high-angle normal
faults, which are common in the axial graben of the rift floor. The fractures apparently served
as conduits for a series of Quaternary volcanoes of basaltic to silicic composition (Ebinger
and Sleep, 1998).
OVC is a classical example of such a Quaternary volcano of the KRS. Characteristic
volcanism within the OVC and other centres exposed on the periphery of the complex (e.g.,
Oserian, Kibikoni, Olenguruoni, Broad Acres and the south-westerly Arcuate Domes)
(Figure 8) are dominated in outcrops by peralkaline rhyolite domes and lava fields. The range
of the stratigraphic units observed in each centre is variable. Similarly, each eruption centre
corresponds to a different age as discussed by Marshall et al. (2009). Two main magmatic
differentiation processes, namely crustal anatexis and crystal fractionation, have been
proposed by Davies and Macdonald (1987) to explain the generation of felsic rocks at OVC
and along the Gregory Rift at large.
2.2 Geology of Olkaria Volcanic Complex
Knowledge of the sequence, nature and architecture of the stratigraphy is fundamental for
understanding both the surface and sub-surface geology of the OVC. The current knowledge
10
on the geology of OVC and its surroundings presented hereafter, derives from the detailed
pioneering work of many scholars (e.g. Clarke et. al., 1990; Marshall et al., 2009; Omenda,
1998). Their work has provided the foundation upon which more recent work has continued
to build and has been refined through subsequent field development. The OVC constitutes
part of the Central Kenya Peralkaline Province (CKPP) (Macdonald and Scaillet, 2006),
which coincides with the Kenya Dome; a ~ 300-400 km wide topographic culmination in
the course of the rift, with the apical region postulated to be lying around the Lake Nakuru
area (Figure 1) (Macdonald and Scaillet, 2006). Nearly 80 volcanic centres composed of
lavas and/or pyroclastic rocks formed by eruptions through fault zones or occuring as steep-
sided domes are found within the volcanic centre, hence the term complex. Clarke and co-
workers (1990) have notably placed the evolution of the OVC to be in the period between
22-20 ka BP.
2.2.1 Surface geology
The surface lithologic units in OVC are dominated by comenditic lavas, pumice fall and
pyroclastics (Figure 6). A large fraction of the pumice fall and pyroclastic deposits is
hypothesized to have originated from Longonot and Suswa volcanoes, lying immediately 20
km east and 40 km south of OVC, respectively. However, there is need for geochemical
analysis of the deposits from the three volcanic centres to establish this assertion. Even
though it has not been possible to enumerate systematically the contribution of pyroclastic
deposits from each of the three centres, the pyroclastic activity at Longonot is presumed to
post-date volcanism at Olkaria (Omenda, 1998). Occurence of the comendites is restricted
to OVC and it is the only area in the whole of the KRS with these rocks on the surface.
Figure 6: Geological map of the Olkaria Volcanic Complex and the surrounding
area (modified from Clarke et al., 1990).
11
Clarke et al. (1990), listed six stages in the evolution of the complex based on their geological
work. A tabulation of the sequence of volcanic groups is provided in Figure 7 and their
distribution summarised in ascending order as discussed below. On the other hand, Olkaria
eruptive centres and their manifestation as distinct topographical units are shown in Figure
8.
Stage 1: The earliest stage, of uncertain age, resulted in a pile of dominantly trachytic lava
and pumice. These are termed Olkaria Trachyte Formation (Ot) and Maiella Pumice
Formation (Mp). The former is exposed mostly along ridges and gullies in the southwest
portion of the complex, whereas the latter is thought to have been erupted from the extensive
vents in the western segment of the complex.
Stage 2: This stage, also of uncertain age is believed to mark the collapse of the proposed
caldera structure, leaving behind a depression 11 × 7.5 km across. The suggested caldera
formation was preceded by an eruption of a large volume of densely welded pantellerites of
the Ol-Njorowa Pantellerite Formation (O1). The latter are mainly exposed in the deepest
sections of Ol-Njorowa gorge (Figures 6 and 7). Based on field observations, Clarke et al.
(1990) relied on a group of coalesced rhyolitic domes (Figure 8), forming distinct
topographic features in the eastern and southern parts of the complex to infer the ring
structure. However, the caldera structure is still poorly constrained because the curved
lineament of rhyolitic domes is adjudged to demarcate only 30% of a caldera rim.
Stage 3: This stage was characterised by early post-caldera activity and is represented by the
eruption of pyroclastic rocks (Op2) and peralkaline rhyolitic lavas (O2) of the Lower
Comendite Member of the Olkaria Comendite Formation. The formation has been dated by
Clarke et al. (1990) at > 9150±110 BP.
Stage 4: Involved ring dome building (O3) and the eruption of thick surge deposits (Op3)
representing the Middle Comendite Member. 14C dating places the age for this member at
between 9150±110 BP and 3280±120 BP (Heumann and Davies, 2002).
12
Stage 5: General resurgence of the caldera floor lead to the formation of short, thick
comendite flows of the Upper Comendite Member (O4). This has been dated at >3280 BP
(Clarke et al., 1990).
Stage 6: This stage was centred on the N-S Ololbutot fissure system (Figure 6). The most
notable expression of this stage is very thick comenditic flows of the Ololbutot Comendite
Member (O5). The youngest of these flows is the Ololbutot lava, dated at 180±50 BP by 14C
(Macdonald and Scaillet, 2006).
Apart from the stages of evolution of the OVC, Clarke et al. (1990) also recognized a series
of eruptive mafic rocks, occuring both to the north and south of the Olkaria Domes field.
The Lolonito Basalt Formation (Ba1) is found to the south (Figure 7), forming a string of
small outcrops and largely comprised of pyroclastic cones and flows. The age of this
formation has been estimated at < 0.45 Ma and it typifies an early phase of the younger Akira
Basalt Formation (Ba2). The Tandamara Trachyte Formation (Tt), lying in the immediate
vicinity of Akira Basalt Formation to the southeast, is postulated to be approximately coeval
with the Akira Basalt Formation (Clarke et al., 1990). To the north of the domes field, there
is widespread mafic volcanism associated with rhyolite emplacement on the expansive
Ndabibi plains (Figure 7). The latter is a low-lying area between the OVC and the Eburru
Volcanic Centre. These have given rise to the Ndabibi Basalt Formation (Bn), comprising
primarily faulted lavas and pyroclastic cones of pre-caldera age. The age range for the Bn is
Figure 7: Surface formations and stratigraphic column of Olkaria Volcanic
Complex. Also shown are a series of eruptive peripheral mafic and trachyte
formations (Macdonald et al., 2008; Marshall et al., 2009).
13
uncertain, a fact attributed to some lavas having well-preserved topography and lacking
vegetation, possibly suggesting that they are young. Nonetheless, some of the Bn flows are
intersected by minor faults and therefore considered older.
2.2.2 Sub-surface geology
Drill cuttings from several geothermal wells have contributed valuable information to the
understanding of the sub-surface strata of the Olkaria geothermal field. In particular, drilling
to depths > 2000 m has provided new insights into the deep sub-surface stratigraphy of the
Olkaria geothermal field by enabling detailed geological analysis of drill cuttings. The litho-
stratigraphic units of the Olkaria geothermal area, as revealed by data from geothermal wells
Figure 8: Geographical extents and stratigraphic units of the Olkaria
eruptive centres. The Ol-Njorowa Pantellerite Formation (O1) is not
shown (Marshall et al., 2009).
14
and regional geology, have been classified into six distinct groups. These groups are based
on age and lithology as discussed in the following section (from the youngest to the oldest).
The Upper Olkaria Volcanics are stratigraphically the youngest unit within the complex
(Omenda, 1998), dated at < 0.95 Ma. Pyroclastics and comendites are prevalent within this
series with minor interbeds of trachyte and basalts (Figure 9). Rocks between the surface
and ~ 1500 m.a.s.l are part of this formation, the youngest being the aforementioned
Ololbutot lava flow dated at 180±50 yrs ago (Macdonald and Scaillet, 2006). Moreover, the
comendites forming a ring of domes east of the complex and those exposed along the Ol-
Njorowa gorge belong to this series. Nonetheless, the comenditic domes have been masked
by the comparatively thick pyroclastic flows deposited in the Domes field and surrounding
area. Some of the deposits are believed to have originated from the adjacent Longonot and
Suswa volcanoes.
The Upper Olkaria
Volcanics formation is
underlain by Olkaria basalts.
Basaltic lavas are
predominant, though
alternating with thin
horizons of tuffs, minor
trachytes and sporadic
rhyolites. The formation is
extensive in the East,
Northeast, Southeast and
Domes fields but absent in
the West field. The
formation has been
intersected by wells at
between 1900 and 1500
m.a.s.l (e.g. Okoo, 2013)
and forms the cap rock for
the Olkaria geothermal
system. Besides, the
formation has been used as a marker horizon within the geothermal field, especially in the
Domes field. The Olkaria basalts are preceded by Plateau Trachytes of Pleistocene age. The
trachytes form part of the Kenya rift floor fissure flows which are exposed to the south and
north of OVC. The formation is characterized by trachyte as the principal rock type with
localized occurence of tuffs, rhyolites and basalts. Plateau trachytes is the reservoir rock for
the East, Northeast, Southeast and Domes fields. Its thickness has been estimated to be
greater than 1.5 km based on borehole data from these fields (e.g., Mwangi, 2012, Okoo,
2013).
The Mau tuffs, dated between 3.4-4.5 Ma, are the oldest rocks encountered within the
complex and are of unknown thickness. The formation has been penetrated by drill holes in
the west fields (Northwest and Southwest), where it forms the reservoir rock. However, it
has not yet been intersected by wells in the other Olkaria fields. The lithotypes constituting
the Mau tuffs are composed almost exclusively of tuffs with minor interbeds of rhyolites,
basalts and trachytes. Pre-Mau volcanics, dated at > 4.5 Ma constitute the fifth series and
Figure 9: Sub-surface lithotypes of the Olkaria Volcanic
Complex (Modified from Omenda, 1998).
15
overlie the basement rock. The lithotypes in this series consist notably of phonolites, basalts,
trachytes and tuffs. Underlying the Pre-Mau volcanics is a laterally extensive basement
system comprising Proterozoic metamorphic amphibolite grade gneisses and schists,
accompanied by marble and quartzites of the Pan-Africa basement system. This unit has
been dated at > 590 Ma. The depth to the basement has been interpreted by Simiyu and
Keller (2000) to be about 5-6 km in the central Kenya Rift. The apparent absence of these
two units, both on the surface and within the sub-surface geological units of the Olkaria
geothermal system, has been interpreted by Shackleton (1996) to be a consequence of a
major graben trending N-S towards the southern flanks of the rift.
2.3 Structural geology and tectonics of OVC
Several researchers have studied and documented the distribution of geological structures in
the OVC and surrounding areas since 1971 (e.g. Baker and Wohlenberg, 1971; Omenda,
1998). As mentioned earlier, the OVC is situated in the cental sector of the elliptically shaped
Kenya Rift, a region marked by a sharp re-orientation in the rift’s trend from a NNW
alignment, to the north of the volcanic complex, to a more or less SSW direction to the south
of the complex (Figure 1). Some dispute still exists regarding the explanation for the change
in the rift’s trend, extending for nearly 120 km before resuming a SSW course. For instance,
Smith and Mosley (1993) have proposed this bend as being an intersection with a large NW-
striking basement structure, the ‘Aswa lineament’ (Figure 1). The theory of the involvement
of the basement structure has, however, been rejected by, inter alios, Bosworth (1987) who
ascribed the shift in the rift orientation to represent an accommodation zone. The latter is
taken to separate two-half grabens, the southern Magadi-Natron sub-basin and the central
Nakuru-Naivasha sub-basin.
An intimate knowledge of the structural controls and their topological relationships in the
OVC is instrumental in determining the reservoir fluid flow at depth, and consequently, in
well siting. The determination of breccia zones associated with faults that could act as
channels for geothermal fluids is of key importance. According to Steiner (1977), some
faults have proved to be productive and serve as efficient conduits for the ascent of
geothermal fluids, while others are known to be totally unproductive. Geological structures,
especially faults and fractures, in the Olkaria geothermal system obviously play a significant
role by serving as pathways for the geothermal fluids. Hence, they are the primary means
for efficient heat transfer from deep to shallower levels of the system. Additionally, the
structures have apparently displaced some geological units within the geothermal system, as
witnessed, for example, by the presence of the Mau tuffs in Olkaria West field but lacking
in wells in the East field. The structures in OVC are divided into rift faults, ring structure,
Ol-Njorowa gorge, and dyke swarms (Figure 10).
16
The structural domains of the OVC are dominated by NE-SW and NW-SE striking faults,
and subordinate N-S, NNE-SSW and ENE-WSW structural trending patterns (Figure 10).
Other faults of variable orientation, though limited, also occur in the area and are not
included in the structural map. Faults are more conspicuous in the East, Northeast and West
Olkaria fields. However, their presence on the surface in the Domes area is severely limited
by the substantial blanketing of the surface geology by thick, younger pyroclastic deposits
(Otieno et al., 2014). The oldest faults of the complex are the NW-SE (e.g. Gorge farm fault)
and ENE-WSW (e.g. Olkaria fault zone) trending faults (Omenda, 1998), and are associated
with the rift formation (~ 30-45 Ma). The most recent fault systems, i.e. the N-S, NE-SW
and NNE-SSW trending ones, are considered to be of a different structural regime and are
thus interpreted to be associated with later tectonic activities, and perhaps, the proposed
caldera collapse. The most obvious of these structures are the Olkaria fracture and Ololbutot
fissure (Figure 10). The trace of the latter is revealed by extensive hydrothermal
manifestations (mainly fumarolic activity), characterized by deposits of kaolinite and silica.
This fault has obviously an impact on the vertical permeability witnessed along the N-S trend
(Mungania, 1999).
Surface geothermal activity in Olkaria coincides with fault arrays. Prominent manifestations,
notably, hot springs, fumaroles and altered grounds, display a strong relationship with the
tectonic lineaments in the area. The Ol-Njorowa gorge, an erosional channel extending for
~16 km in total length, has been dated at about 9 ka BP (Clarke et. al., 1990). It is aligned
roughly in a NE-SW direction and marks the boundary between the East and the Domes
Figure 10: Structural map of the Olkaria Volcanic Complex displaying the orientation
of various fault patterns (modified from Omenda, 1998).
17
fields (Figure 10). Studies by Clack and co-workers (1990) suggest the gorge to be a product
of faulting and was later deepened by catastrophic southward overflow of Lake Naivasha
when the lake level was much higher than at present. Numerous felsic dykes trending NNE-
SSW and volcanic plugs (necks) are widely distributed along the gorge, strongly suggesting
that faulting played a key role in its initiation (Otieno et al., 2014). Further evidence is
provided by the active, boiling fumaroles, possibly attesting to reactivation of some of the
faults along the gorge.
The arcuate alignment of rhyolitic domes in the eastern and south-eastern parts of Olkaria
has been interpreted by Omenda (1998) to indicate the presence of a buried caldera. In his
hypothesis, he proposed the arcuate alignment of rhyolitic domes and ridges to be products
of a resurgence. Perhaps a more plausible explanation is that the ring structure resulted from
magmatic stresses in the Olkaria magma chamber (Omenda, 1998). Nonetheless, there is no
overwhelming evidence for any of the many possible explanations. As highlighted earlier,
the exposure of geologic surface structures is limited in the Domes area and its immediate
vicinity due to the cover of the surface by pyroclastic deposits. Most of the structures in this
area are sub-surface and have been deduced through surface mapping along the deep gullies
and well drilling operations. Evidence for buried faults, has been found through stratigraphic
correlation of wells (e.g. Lagat, 2004) owing to the presence of excellent marker beds (Figure
9). Cave-ins and total loss of circulation returns during drilling operations, for instance in
well OW-915B (KenGen, 2012), have also served as a clear indicator for the existence of
sub-surface fault structures.
2.4 Hydrogeological setting
The hydrogeology of OVC, and to a greater extent the surrounding volcanic centres (i.e.
Eburru, Longonot and Suswa) within the central sector of the KRS, is controlled primarily
by the rift faults. Tectonic movements of the rift undoubtedly played a pivotal role both on
a large-scale, by forming the regional faults, and on a small-scale, by creating the local
fracture systems of enhanced permeability. These faults and fracture systems act as
permeable pathways by channelling lateral groundwater flow from the recharge areas into
the low-lying geothermal systems along the rift axis. Ordinarily, faults have the capacity to
propagate fluid flow by providing zones of high permeability. Likewise, faults may
simultaneously offset zones of relatively high permeability (Allen et al., 1989), in which
case they cause compartmentalization of lateral flow.
Recharge into the geothermal field at Olkaria is exclusively of meteoric origin (Ojiambo and
Lyons, 1993), derived from high altitude precipitation at the rift escarpments (Mau
escarpment to the west and Kinangop plateau to the east) (Figure 11). The two highlands
have a maximum elevation of 3080 m and 2740 m, respectively, and experience an annual
rainfall of between 1200-1500 mm. On the other hand, Olkaria geothermal field and its
environs receive an estimated average annual rainfall of about 826 mm (Allen et al., 1989).
The structures controlling deep recharge within the system are the NW-SE and NNW-SSE
trending faults (Figure 10). As previously reported (Ambusso and Ouma, 1991), the
rejuvenated pre-Pleistocene, ENE-WSW trending Olkaria fault zone (Figure 10) is a major
structural divide transecting the Northeast and West fields. Therefore, it is regarded as an
important structure forming a hydrologic divide within the geothermal system. This, for
instance, has been demonstrated by the presence of a liquid-dominated reservoir without a
18
steam cap in the north and a liquid dominated two-phase reservoir overlain by a steam cap
in the south of the fault (Ambusso and Ouma, 1991). Some water from the recharge areas
also enters the Olkaria geothermal system through the late-Pleistocene, relatively young,
tectonically reactivated N-S faults and fractures (e.g. Ololbutot fault and Olkaria fracture
zone, respectively). These structures control the axial groundwater movement through the
geothermal system, though have a shallower influence in comparison to the major rift faults
that offer deep recharge (Omenda, 1998).
Other features which have been noted to enhance fluid flow within the Olkaria system, and,
to a larger extent in the Domes area, are lithological interfaces and, jointed and fractured
lavas. Similarly, stable isotope studies have been employed (e.g. Nkapiani, 2014) to
determine the origin of thermal waters and the reservoir recharge areas of the Olkaria
geothermal system. Based on hydrogen isotope studies, the reservoir fluid isotope
composition of the thermal waters indicate two potential recharge zones for the system; the
eastern rift flank (δD = ~ -24‰) and the western rift flank (δD = ~ -30‰). Results of
hydrogen isotope determinations suggest the West, East and Northeast fields to be recharged
by waters from the western escarpment. The Domes field obtains recharge largely from the
eastern rift wall (Nkapiani, 2014). Whether or not sub-surface drainage from Lake Naivasha
Figure 11: Piezometric map showing reservoir recharge areas
for Olkaria and other surrounding geothermal fields along the
rift axis (modified from Allen et al., 1990).
19
contributes to recharge of the Olkaria geothermal system has intrigued many researchers and
is still a subject of debate (e.g. Allen et al., 1989).
2.5 Reservoir properties of Olkaria field
In this section, attention is focused principally on the most important features of the reservoir
system of the Olkaria geothermal field. A comprehensive review of all the aspects of
reservoir characteristics of the Olkaria geothermal field is beyond the scope of a single sub-
section. The Olkaria geothermal reservoir has developed as a consequence of the OVC.
Presently, the geothermal system supports ~ 654 MWe (installed capacity) through power
stations operated by KenGen and Ormat Technologies (OrPower) (Table 1). It is a two-phase
liquid dominated system overlain by a comparatively thin steam-dominated zone, ranging
from 100 to about 200 m in thickness at 240oC. The hydrogeological features of the field are
distinguished by four up-flow zones in the general area. Extensive fluid circulation in the
field is controlled by major faults and fracture zones as discussed earlier. The ultimate heat
source of the system has been attributed to a deep-seated magma chamber, which peaks in
several locations in the form of three main intrusions, possibly partially molten. Based on
seismic studies by Simiyu and Keller (2000), the intrusions extend to shallower levels,
approximately 6-8 km, and are suggested to lie beneath Olkaria Hill, Gorge Farm volcanic
centre and in the Domes area. Temperature and pressure models supports this observation
and, have identified four main up-flow zones related to these heat sources (Figure 12)
(Mannvit consortium, 2011).
Figure 12: A vertical schematic sketch of the conceptual model of
Olkaria geothermal system. The suggested heat sources, up-flow zones
and temperature cross-section are shown. The green arrow is directed
north (Mannvit consortium, 2011).
20
The first up-flow zone feeds the Olkaria west fields and is linked to the Olkaria Hill (Figure
2) heat source body. The second up-flow zone, feeding the Northeast and East fields as well
as the northwest corner of the Domes field, is connected to the heat source beneath the Gorge
Farm volcanic centre (Figure 8). The final up-flow zone, associated with the ring structure,
is visible in the southeast corner of the Domes field and is linked to the heat source suggested
beneath the area. Reservoir chemistry data support the existence of the up-flow zones as
defined by relatively high Cl- concentrations and high Na/K temperatures within these zones
(Mannvit consortium, 2011). Above the steam zone, basaltic rocks, and to a lesser extent
trachytes, form an impermeable caprock, subsequently marking the top of the reservoir. This
proposition has been supported by reservoir modeling (Ambusso and Ouma, 1991), which
reveals sharp temperature increases below this formation. The depth to the caprock,
however, varies from one sub-field to another, but in many instances lies between 1900 and
1300 m.a.s.l, as revealed by several stratigraphic correlations (e.g. Okoo, 2013; Mwania et
al., 2013). The bulk of the reservoir rocks, save for in the West fields, are trachytes of
Pleistocene age with well developed fractures. The vertical extent of the reservoir is quite
uncertain and varies from one sub-field to another. Ambusso and Ouma (1991) have
suggested it to be of the order of several hundred meters. However, the presence of extensive
intrusive bodies (e.g. syenite, granite, etc) at greater depths (e.g. below 2500 m) probably
indicates proximity to the bottom of the reservoir.
2.6 Summary of drilling history
Well OW-922 is a vertical well drilled to a total depth of 2990 m from the cellar top (CT),
to the east at approximately 2.5 km from the proposed ring structure. The well was spudded
on 16th of April and completed on 19th of September, 2014, after a period of 156 days. This
translates to 101 more days than planned, since the well had initially been expected to be
completed on 12th June, 2014. The drilling stages comprised the surface, intermediate and
production hole as well as a production section. Delays in the drilling were occasioned by
various challenges as outlined in the description of different phases. Casing depths were
measured from the flange. A summary of the drilling phases and casing design of well OW-
922 is illustrated in Table 2 and the drilling progress is shown in Figure 13.
Table 2: Drilling phases and casing design for well OW-922.
Drill Rig
Stage
Hole
diameter (")
Depth (m) Casing
From To Casing
diameter (")
Type
GWDC-116
Phase 1
Phase 2
Phase 3
Phase 4
26
171/2
121/4
81/2
0
58
300
1200
58
300
1200
2990
20
133/8
95/8
7
Surface
Anchor
Production
Slotted
liners
21
2.6.1 Phase 1: Drilling of the Surface hole
Surface hole was drilled with a 26" roller cone milled tooth bit using high-viscosity bentonite
based-mud to a depth of 58 m rotary kelly bottom (RKB). Full circulation returns occured at
the surface throughout the drilling of this phase. The well was circulated to the bottom and
no obstructions were detected. It was cased using a 20" outside diameter (OD) surface
casing. Casing shoe was set at 57.1 m RKB and the casing column successfully cemented.
This phase lasted 4 days.
2.6.2 Phase 2: Drilling of the Intermediate hole
Drilling of phase 2 started on 20th April, 2014 using a 171/2" roller cone bit with aerated
water and foam as the drilling fluid. Like during the drilling of the surface hole, full
circulation returns were received on the surface down to the section total depth (TD) of 300
m RKB. The hole was circulated before the 133/8" anchor casing was run in hole (RIH) and
casing shoe set at 299.19 m RKB. The casing column was cemented in place in three stages;
the primary cementing and two backfills. Drilling of this phase was completed in 6 days.
Figure 13: Drilling progress curve for well OW-922 (KenGen,
2014).
22
2.6.3 Phase 3: Drilling of Production hole
Drilling of this phase began on 26th April, 2014 with the use of a 121/4" roller cone bit.
Drilling progressed reasonably well with aerated water and foam being used with good
circulation returns realized at the surface. However, several unusual events occurred. On 29th
April, the drill string got stuck at a depth of 603 m RKB. Several attempts to free the string,
including pumping a mixture of PIPE-LAX and diesel, proved successful after 2 days. A
decision was then made to pull out of hole (POOH) and inspect the drill strings, before RIH
to continue drilling. Further attempts to proceed with drilling proved futile due to a second
obstruction encountered at 520 m RKB. At this point in time, the well condition had become
complicated. The crew embarked on reaming the wellbore from surface to bottom with
aerated water and foam, receiving no circulation returns at the surface. This was done in an
attempt to clear any obstructions along the well course.
While reaming almost reaching the bottom (603 m RKB), the drill string got stuck once
again. Efforts to free the string were immediately initiated. This was on 1st May, 2014. To
overcome this impediment, a decision was made to carry out a plug job. A total of 24 plug
jobs were conducted around this section. The decision proved successful and the section was
finally by-passed. Normal drilling commenced again on 29th May, with the use of mud and
progressed smoothly to TD of 1200 m RKB. Circulation of the wellbore and a wiper trip
were undertaken without any considerable challenges. RIH of the 95/8" OD production
casings followed and the casing bottom was set at 1197.86 m RKB. The casing column was
then cemented in place in three stages; the primary cementing and two backfills. Drilling of
this phase lasted 55 days.
2.6.4 Phase 4: Drilling of Production section (Main hole)
Drilling of the production section resumed on 20th June, 2014 with the use of a 81/2" roller
cone bit, with aerated water and foam as the drilling fluid, resulting in intermittent circulation
returns. Drilling progressed without any major hindrance before an obstruction was
encountered at 1359 m RKB. This was successfully overcome after 1 day of reaming, and
drilling continued smoothly with excellent circulation returns. On 8th July, (84th day) a tight
hole was encountered at 1280 m RKB while RIH with a new drill bit. A decision was made
to ream the wellbore with aerated water and foam before resumption of drilling, which
proved fruitful. The final depth of 3000 m RKB was achieved on 13th July, 2014 (89th day).
The crew embarked on circulation to unload the wellbore before RIH hole the 7" perforated
liner. However, an obstruction was once again encountered at 1270 m RKB while conducting
wiper trip prior to POOH and further reaming followed. A possible reason for the
development of obstruction between about 1200-1360 m RKB was thought to be the
presence of swelling clays. XRD analysis (Appendix B) confirms the occurence of swelling
clays around this section. After a wide consultation, a decision was made to pump a mixture
of 1.75 tonnes of sodium hexametaphosphate [(NaPO3)6] (a dispersant) and 10m3 of water
into the wellbore to help break the clay bonds. This proved successful. Eventually, the
wellbore was successfully reamed to the bottom using aerated water and foam, receiving full
circulation returns at the surface. A wiper trip was conducted and the drill string pulled out
of the hole before the 7" slotted liner and 2 plain liner pipes were RIH and squatted at the
bottom. This phase lasted 91 days.
23
3 Sampling and Analytical Methods
3.1 Sampling
Studies of various parameters in borehole geology require comprehensive and accurate
sampling. The information obtained depends on the samples quality. Consequently, a
systematic selection of samples is indispensable. Principally, this study involves sampling
of drill cuttings (chips), measurements at discrete depths and laboratory analyses. Drill
cuttings analysis is basically a visual method of semi-quantitatively describing the host rock
characteristics from drill cuttings obtained during a drilling operation. The analyses are
crucial in constraining the sub-surface conditions of the geothermal system (Reyes, 2000).
These encompass, but are not limited to, lithostratigraphy, hydrothermal alteration rank and
intensity, permeability, petrochemical characteristics and zonation patterns.
Unlike in the case of core samples, interpretation based on drill cuttings analysis is not
without a fair share of its challenges, including imprecise sample location as a consequence
of relatively large lag times. In other words, drill cuttings do not represent exact horizons,
but a limited depth range. The time taken by drill cuttings to get to the surface is dependent
on the type of circulation fluid, drillhole diameter, flow rate and circulation loss. In addition,
chances of collecting unrepresentative samples caused by sample mixing as a result of
drilling through soft formation are quite high. Further, grinding of cuttings to extremely fine
grains (almost powdery) by the drill bit leads to loss of essential geological features. No
cores were collected in well OW-922. Therefore, all the descriptions and interpretations in
the present study are based solely on the analysis of drill cuttings. Uncertainty in sample
depth does not substantially affect the conclusions reached in this study. Cuttings were
collected at 2 m interval, except in cases of partial circulation returns. In such cases, it was
carried out at 4 m interval. The sampling intervals along with total loss zone depths were
constantly recorded in the rig geology notebook.
3.2 Analytical methods
To achieve the specific objectives of this study, several analytical techniques have been
employed as discussed in a following section. The procedure and a summary of the analytical
results obtained are presented in different appendices.
3.2.1 Binocular microscope analysis
The main parameters described by binocular microscope analysis include colour of the rock,
grain size, rock fabrics, primary/original mineralogy, hydrothermal alteration rank and
intensity, rock type and features that point to the possible occurence of feed zones or
intrusions. Samples were scooped up using a petri dish and washed to enhance visibility and
reveal any obscured features. Binocular microscope analysis for this study was carried out
at Iceland Geosurvey (ÍSOR) petrography laboratory using an Olympus, SZX12 binocular
microscope. A detailed description of the lithology is presented in Appendix A.
24
3.2.2 Petrographic microscope analysis
With this analytical technique one can ascertain in much finer details the rock type and
hydrothermal alteration grade based on the optical properties of minerals. Hydrothermal
alteration intensity, paragenetic sequence and some other features not obviously discernible
under the binocular microscope can be described in detail. A total of 70 thin sections were
analysed at the University of Iceland petrographic laboratory, using, an Olympus BX51 and
Leitz Wetzlar petrographic microscopes with a magnification range of between 4x and 50x.
3.2.3 X-ray diffraction analysis
The X-ray diffraction (XRD) analysis technique is concerned primarily with structural
aspects of clay minerals and other minerals (e.g. zeolites, amphiboles, etc). The
diffractometry analysis is based on the principles of Bragg’s law. When a beam of X-rays is
passed through a particular mineral and certain geometric conditions are achieved (Yoshio
et al., 2011), the X-rays scattered from the crystalline sample can undergo constructive
interference, thus producing a diffraction peak. A diffraction pattern (made by the
diffractometer) records the X-ray intensity as a function of 2𝜃 (incident angle of the X-rays)
which is characteristic of a specific mineral. In this study, a total of 105 samples were
analysed exclusively for clay minerals at the KenGen XRD laboratory. A summary of the
XRD analytical results are presented in Appendix B.
3.2.4 Inductively Coupled Plasma-Optical Emission
Spectrometry (ICP-OES) analysis
The ICP-OES method has today become a valuable tool for quantitative elemental analysis.
The basic principle behind the method is that atoms of an analyte emit electromagnetic
radiation at certain characteristic wavelengths when atomized (Rollinson, 1995). The
wavelengths of the light emitted are then separated and measured in a spectrometer. This
yields an intensity measurement that can be easily converted to an elemental concentration
by comparison with calibrated reference standards (Rollinson, 1995). Actual analysis
involves spraying the sample solution into the core of an inductively coupled argon plasma,
generating temperatures of approximately 7,000-10,000 K. At such high temperatures, all
analyte species are atomized, ionized and thermally excited, and can then be detected and
quantified with an optical emission spectrometer (OES). The extremely high temperature
ensures that the sample matrix does not produce any interfering components that might
interfere with the analysis.
For this study, a total of thirty homogeneous samples from the study well were analysed for
major and trace elements using the Spectro Ciros 500 ICP-OES equipment at the Faculty of
Earth Sciences, University of Iceland. The major element oxides analysed are SiO2, Al2O3,
FeO, MnO, MgO, CaO, Na2O, K2O, TiO2 and P2O5. Trace elements analysed include, Ba,
Co, Cr, Cu, La, Ni, Rb, Sc, Sr, V, Y, Zn and Zr. Samples were analysed using the following
USGS internationally recognized standards and error margins; (1) Basalts: multiple
measurements of standard Basalt, Hawaiian Volcano Observatory (BHVO)-1 2σ (external
reproducibility). Analysis for Si, Al, Fe, Mn, Mg, Ca, Na and P was done with 2σ <1.8%. The error margin for trace elements in basalts ranged between 4-8%, (2) Rhyolites:
measurements of standard Rhyolite, Glass Mountain (RGM)-1. Elements were analysed with
25
2σ values as follows; Si and Mg ~ 1%, Ti <2%, Al, Ca, Fe, Na and K <4-6%, Mn and P
<8%. The error margin for trace elements in rhyolites is the same as for basalt. A detailed
description of the ICP-OES procedure for sample preparation, instrument set-up and data
processing, is presented in Appendix C. Full analytical results are listed in Table 5.
3.2.5 Electron microprobe analysis
Electron microprobe (EMP) analysis is routinely used for characterizing the compositions
and compositional variations of minerals within rock samples. The basic theory of EMP
analysis relies on the production of electrons and X-rays during the interaction of the incident
electron beam with the specimen (Schiffman and Roeske, 2014). Consequently, with the
electron beam focused on a small area on a sample surface, characteristic X-rays are
generated for qualitative or quantitative analysis at micrometer-scale spatial resolution. The
volume of the electron-matter interaction varies directly with the accelerating potential of
the electron beam (Schiffman and Roeske, 2014) and inversely with the mean density of the
sample. Two key processes, namely, the generation of backscattered electrons and the
generation of characteristic X-rays, are responsible for producing the main signals utilized
in EMP analysis.
A total of 10 samples from the four wells (OW-905A, 910, 917 and 922) were selected for
analysis of secondary minerals of interest. These represent four samples containing epidote
and six samples containing chlorite. The cuttings were subsequently moulded in epoxy,
polished, carbon coated and analyzed using a JEOL JXA-8230 electron microprobe located
at the Institute of Earth Sciences, University of Iceland. Elements analysed and the
conditions under which the probe was run are described in section 4.9. Prior to quantitative
analysis with the electron microprobe, the minerals of interest, namely, chlorite and epidote,
were first studied with a Hitachi TM 3000 Scanning Electron Microscope (SEM). Results
for EMP analysis for chlorite and epidote are presented in Appendix D.
27
4 Results
4.1 Lithology of well OW-922
Knowledge of the sequence, nature and architecture of the stratigraphy is of paramount
importance for understanding the geothermal system around well OW-922. Analyses based
on binocular and, petrographic observations and ICP-OES have revealed eight types of
lithologies intersected by well OW-922. These comprise pyroclastics, tuffs, rhyolite, basalt,
basaltic trachyandesite, trachyte, trachydacite and trachyandesite. No intrusions were
intersected by the well. A description of all lithological units encountered in the well are
discussed below and presented in Figure 14. However, a detailed description of the various
units and intervals of occurence along the well profile is presented in Appendix A.
4.1.1 Pyroclastics
The pyroclastic formation is a member of the Upper Olkaria Volcanics according to the OVC
stratigraphical classification scheme. The unit occupies the uppermost 302 m of the well
column.The formation generally varies in colour from light grey to yellowish-brown,
comparatively heterogeneous and indurated deposits dominated by ash-sized particles,
volcanic glass, scoria, highly vesicular pumiceous and tuffaceous fragments. It is oxidized
in all cases. The yellowish-brown unit are pyroclastics from Olkaria, whereas the light grey
to grey in colour counterparts are pyroclastics derived from Longonot volcano. Longonot
pyroclastics extend from the surface to ~ 96 m depth, whereas, Olkaria pyroclastics, derived
from the early post-caldera activity (Op2) are observed between ~ 190 and 302 m depth.
4.1.2 Tuffs
Tuffs in the well occur predominantly as interbeds and intercalations between lava flows,
marking different eruption episodes. The units range in thickness between 2 and 52 m. Two
distinct tuff varieties have been recognized in the well, lithic and glassy (vitric) types. The
former is the dominant, varying in colour from light grey, grey to brown and, occasionally,
pale green. The latter is due to alteration. Trachytic and basaltic fragments are the most
common crystalline fragments, and are embedded within the lithic matrix. The phenocryst
phases observed in optical petrography are quartz, sanidine and augite. Moderate to high
vesicularity is a common feature of these tuffs, with majority of the vesicles being irregular
in shape. Vesicles are filled by alteration minerals, primarily zeolites and chalcedony. Lithic
tuff was noted occuring between 302-350 m, 362-364 m, 408-422 m, 476-498 m, 564-582
m, 726-744 m, 952-966 m, 1440-1600 m and 1600-1642 m. Vitric tuff, on the other hand,
was only observed between 1260-1312 m. The vitric tuff is white in colour, and has a glassy
matrix as observed under the petrographic microscope, with minor sanidine and quartz
phenocrysts. Generally, the tuffs display moderate to high intensity of alteration, particularly
in the basal rubble (contacts).
28
4.1.3 Rhyolite
The rhyolite encountered in this well is of granular type and ranges in colour from light grey
to grey to pink. It is massive, holocrystalline, and generally sparsely to moderately phyric.
Also, it is hard, probably due to intense silicification. The matrix has commonly quartz
porphyries and subordinate sanidine phenocrysts. Small mafic mineral grains are
disseminated in the aphanitic groundmass and have been proved through optical petrography
to be riebeckite and oxides, probably magnetite. The formation is not extensive in the study
well as compared to that observed in wells drilled along and within the proposed caldera
structure. It is restricted to depths between 1046-1150 m, 1214-1224 m, and 2370-2592 m.
Based on petrochemical data, the rhyolite is silica-rich with SiO2 content ranging between
69 and 73 wt.%.
4.1.4 Basalt
The formation ranges in colour from dark grey to brown, with the brown varieties depicting
high intensity of hydrothermal alteration. The dark grey basalt is fine grained and
holocrystalline as observed between 498-504 m, 1198-1214 m and 1642-1654 m. In addition,
it is dense and the predominant phenocryst phases are plagioclase and clinopyroxene
(notably augite) as well as a large abundance of opaque oxides, mainly titanomagnetite.
Olivine phenocrysts are rare and their prescence is disclosed by crystal outlines of the
alteration products, e.g. calcite and chlorite. Plagioclase laths are also widespread in the
holocrystalline matrix and outweigh clinopyroxene in all cases. On the other hand, vigorous
alteration has completely obliterated the primary mineral phases and original texture in the
brown coloured basalt. The latter occurs between 1150-1198 m. Nevertheless, vesicles are
observed largely filled with calcite and variably green clays in both cases.
4.1.5 Basaltic trachyandesite
This rock type of intermediate composition is scarce in the well, and generally along the
EARS system based on petrochemical data from previous studies (e.g. Peccerilo et al., 2007).
It was identified between 1654 and 1700 m through petrochemical analysis. The rock is
mostly brown to grey (probably as a result of intense alteration), though it occasionally
appears light grey (e.g. between 1662-1666 m and 1680-1682 m). It is aphyric to slightly
phyric with minor plagioclase phenocrysts. Other phenocryst phases observed are
amphiboles and clinopyroxenes. The latter occur characteristically as slender green needles.
Mafic silicates in most cases (~ 80%) display more or less complete replacement by brown
clays.
4.1.6 Trachyandesite
Like basaltic trachyandesite, this rock type is also scarce in this well and was confused for
trachyte during the initial binocular analysis. The unit was identified through petrochemical
analysis as occuring only between 2780-2784 m. The rock has a composition intermediate
between trachyte and andesite and represents less fractionated magma. The notable
phenocryst phases observed in thin section are plagioclase, sanidine and pyroxenes, mainly
aegerine-augite.
29
4.1.7 Trachydacite
This unit was identified between 1700-1730 m, 2186-2210 m, 2326-2370 m and 2592-2780
m through petrochemical analysis. The rock has a composition intermediate between
trachyte and dacite. The colour varies from pale grey to grey, to pink. It is aphyric to poorly
phyric, and contains clinopyroxene and sanidine as the principal phenocrysts. Vesicles are
rather scarce, but where present are filled by calcite, chlorite and secondary quartz.
4.1.8 Trachyte
This is the dominant lithology in the well and the main sequence is ~ 1200 m thick, as
observed between 1700-2900 m. It is the reservoir rock of the geothermal system around
well OW-922. This unit is predominantly grey, though colour ranging from pale green to
brown is also noticed. In most cases, the unit is phyric with sanidine, subordinate feldspars
and quartz phenocrysts as well as specks of mafic minerals such as pyroxenes and
amphiboles. The pyroxenes are in many cases aegerine-augite while the dominant amphibole
is dark blue riebeckite, recognised through petrographic analysis. Occasionally, the
formation is observed mixed with basaltic and tuffaceous fragments (e.g. between 1248-
1260 m and 2156-2292 m), suspected to have fallen from above during drilling. Trachytic
texture with flow banding is a characteristic feature of these trachytes with sanidine laths
aligned in a particular direction. In addition, elongated vesicles are abundant, a display of
the direction of lava flow.
4.2 Hydrothermal alteration
Hydrothermal alteration is a metasomatic process involving recrystallization of rock
forming/primary minerals and deposition of secondary minerals in voids, to a new authigenic
mineral assemblage due to interaction with thermal fluids. The new mineral assemblage is
more stable, or at least metastable, at a specific temperature, pressure and fluid composition
range within a particular geothermal system. The stability of the hydrothermal minerals is
determined by their solubility and the rock dissolution process. Hydrothermal alteration
involves interaction of the hot aqueous fluids with the host rock (a process popularly known
as water-rock interaction), resulting in a change both in the composition of the fluids and
the rocks (Brown, 1978). Circulation of thermal fluids through the geothermal system is a
consequence of the heat generated by magma or hot crustal rocks.
During ascent to the surface along permeable structures, the thermal fluids circulating within
the geothermal system directly deposit dissolved mineral constituents in vesicles, vugs or
veins. Correspondingly, the thermal fluids dissolve ions from the host rocks, consequently
forming secondary minerals. By observing alteration overprints or the sequences of
hydrothermal minerals deposited in the veins or vugs, it is possible to unravel the prevailing
reservoir temperature, permeability and fluid composition (Reyes, 2000). According to
Mielke et al. (2015), hydrothermal alteration is defined by both its grade and intensity. The
term grade implies a definite assemblage of hydrothermal minerals present. Intensity of
alteration, on the other hand, defines whether the alteration is incipient, partial, or complete.
Accordingly, alteration is described using qualitative terminology such as low, medium, or
high, depending on if a primary constituent is incipiently, partly, or completely altered in
that order (Reyes, 2000).
30
Formation of hydrothermal minerals is generally dependent on the temperature, pressure,
reservoir permeability, protolith, fluid composition (especially pH), and duration of fluid-
rock interactions (Brown, 1978). Although pressure has little direct effect on hydrothermal
alteration, an important aspect is that it controls the depth at which boiling occurs, and thus
separation of vapour and gases. Some of these factors listed are so intimately interconnected
that, at first sight, it is often impossible to separate one from another (Brown, 1978). Products
of hydrothermal activity in the geothermal system in the vicinty of well OW-922 occur both
as replacement of unstable primary components or glassy matrix of the rocks (Table 3) and
as open space fillings in veins, fractures, and vesicles.
Table 3: Alteration of primary minerals and order of replacement in well OW-922
(adapted from Brown, 1984).
Order of replacement
Su
sceptib
ility
Primary phases Alteration products
Volcanic glass
Olivine
Plagioclase
Pyroxenes, amphiboles
Sanidine
Fe-Ti oxides (opaques)
Clays, calcite
Clays, chlorite, calcite, quartz
Albite, clays, zeolites, calcite
Clays
Adularia, clays, calcite
Titanite, pyrite, haematite
4.2.1 Distribution of hydrothermal alteration minerals
Hydrothermal alteration minerals in well OW-922 were observed to occur both as
replacement of primary minerals, and as deposition as veins and as fracture and vesicle
fillings. The distribution of alteration minerals in well OW-922 was determined with
binocular, petrographic and XRD analyses. The main hydrothermal minerals encountered in
the well include: zeolites (mesolite, scolecite and wairakite), chalcedony, haematite, titanite,
calcite, albite, pyrite, clays (smectite, illite, chlorite and mixed-layer clays (MLCs)), quartz
and epidote. This is presented in Figure 14 and discussed further below.
Zeolites are hydrated aluminosilicate minerals structurally characterized by a framework of
linked tetrahedra, each consisting of four oxygen atoms surrounding a silicon or aluminium
cation. This three-dimensional network has open cavities in the forms of channels and cages
(Wise, 2005), which are occupied by water molecules and non-framework cations.
Occurence of zeolites in geothermal environments is usually typified by three different
fabrics, namely, fracture and vesicle fillings, as replacement products of relict feldspars, and,
as recrystallization products of siliceous or glassy groundmass of volcanic rocks (Wise,
2005). Most zeolites, with the notable exception of wairakite, are low-temperature minerals
which, are thermodynamically unstable at temperatures >110oC.
Unlike in most wells in the Olkaria geothermal system, the occurence of zeolites in this well
is restricted only to two low-temperature types, mesolite, a high-silica zeolite, and scolecite,
a low-silica zeolite. Both were observed largely as vesicle infillings in tuffs between 302-
350 m, 362-364 m and 408-422 m, and in trachyte between 364-408 m. Mesolite was
identified by the occurence of white, thin, needle-like crystals, which radiate away from the
base, forming spiky ends. By contrast, scolecite was in most cases colourless and formed
clusters radiating from a central point of growth.
31
Wairakite, a calcium-rich zeolite, was first observed in a vesicular tuff between 494-496 m.
It occurs in association with hydrothermal quartz and more rarely with calcite. It is
characterizd by a colourless to murky white colour with aggregates of subhedral crystals,
coating fractured surfaces. It is also identified filling out the abundant pore spaces in
pumiceous fragments at 574-578 m. The occurence is nonetheless sporadic. The first
appearence of wairakite at 494 m in well OW-922 signifies a temperature of about 200oC
and it seems to be the only zeolite stable above that particular temperature (Kristmanndóttir,
1979).
32
Figure 14: Lithology, location of feed zones, distribution of alteration minerals and
alteration zones of well OW-922.
33
Chalcedony is a cryptocrystalline variety of quartz, characterized by a translucent white to
grey to light blue, and in some instances shades of brown. The mineral was first observed at
412 m as a white thin-layered lining in a vesicle. It is also found intermittently and occurs in
the deeper sections of the well, for example between 2576-2578 m. Chalcedony forms at
temperatures from 150oC to as high as about 180oC (Kristmanndóttir, 1979), but gradually
recrystallizes to hydrothermal quartz with increasing temperature (e.g. between 2114-2116
m).
Haematite is a low-temperature Fe-oxide whose occurence is associated with the incursion
of cold groundwater and other oxidizing conditions in a geothermal system (Brown, 1978).
The mineral was observed by binocular analysis as a silver-grey, brown to reddish brown in
colour. It was noted in moderate amounts between 1192-1200 m, and occurs irregularly from
2342 to 2982 m.
Titanite is recognized as white coloured mineral, usually as a replacement of the mafic
silicates. The mineral is very common in this well and was first observed between 422-476
m, occuring sporadically to the well bottom. In petrographic analysis, titanite is identified
by its high relief, brown to dark brown colour and shows dusty-cloud discoloration formed
by the alteration of the ferromagnesian minerals.
Calcite is ubiquitous in the well and occurs in all the hydrothermal mineral alteration zones.
The occurence is particularly prevalent between 1120-1730 m and 2210-2990 m, where it is
rated at a scale of between 2 and 3 (Figure 14). The latter indicates high abundance, whereas,
the former signifies moderate abundance. The mineral is noted as a replacement of primary
forming minerals, especially plagioclase and pyroxene phenocrysts, and discrete silicic
volcanic glass. This process inevitably requires introduction of CO32−ions from the
geothermal fluids (Steiner, 1977). In addition, calcite is observed as precipitates in veins and
vesicles (e.g. between 520-522 m, 1600-1646 m) and as platy calcite (e.g. between 1646-
1648 m and 1944-1946 m), possibly indicating deposition through mixing of fluids with
different compositions. However, in some rare cases, it is noted as a complete groundmass
replacement. In active geothermal systems, calcite is usually regarded to be stable over a
broad temperature, ranging from ~ 50oC to about 300oC (Simmons and Christenson, 1994).
Pyrite is a hydrothermal mineral whose presence in significant amounts has traditionally
been associated with permeable zones in most geothermal systems. It was first observed at
302 m in a tuff. Like calcite, pyrite occurs pervasively to the well bottom. The occurence
varies between 1 to 3, implying low and high abundance, in that order. In binocular analysis,
the mineral was recognized by euhedral cubic crystals with a shiny brass yellow lustre,
predominantly disseminated and, in other instances, embedded in the rock matrix. On rare
occasions, for example between 1312-1420 m, the mineral is seen filling veinlets. The cubic
crystal nature was equally applied while distinguishing pyrite from the other opaque
minerals in petrographic analysis. According to Steiner (1977), pyrite is a product of the
replacement of magnetite (equation i) or result from the attack of FeO in ferromagnesian
minerals with H2S gas (equation ii); such processes are known to occur in other geothermal
systems.
(i) Fe3 O4 + 6H2S ↔ 3FeS2 + 4H2O + 2H2
(ii) FeO + 2HsS ↔ FeS2 + H2O + H2
34
Furthermore, the Fe in the pyrite is suggested to be possibly introduced by the geothermal
fluids, which commonly contain minute amounts of Fe (Steiner, 1977). An alternative
interpretation is that the Fe may be leached from neighboring rocks and transported perhaps
only for a short distance before being converted to pyrite. Pyrite also occurs over a broad
temperature, ranging from ~ 50oC to about 300oC (Brown, 1978).
Hydrothermal quartz is observed mainly as an open-space filling mineral in veinlets,
vesicles, vugs and interstices of fractured rocks. It is generally colourless to occasionally
white coloured, hexagonal shaped, commonly found as subhedral to euhedral crystals that
rarely exceed ~ 2 mm in length. The mineral was first noted between 476-478 m filling
vesicles in a tuff, designating formation temperature in excess of 180oC (Reyes, 2000). The
occurence is however, sporadic to the well bottom. In many cases, it is found in association
with wairakite, pyrite and titanite. In thin section, quartz was easily identified by the
undulating extinction, lack of cleavage and low interference colours. Petrographic evidence
unequivocally shows that quartz to a greater extent precipitated or was introduced by
geothermal fluids as opposed to crystallised from SiO2 released during alteration of silicic
glass and/or feldspar phenocrysts and ferromagnesian minerals.
Albite. The term albitisation in this context is reserved for the replacement of primary K-
feldspar (specifically sanidine in this case) and plagioclase phenocrysts by hydrothermal
albite. The upper limit of this mineral in the well is at 802 m and it occurs irregularly but
disappears totally below 952 m depth. The mineral was identified during binocular
investigation by its cloudy white and striated surfaces. The mineral grains vary largely
between anhedral to subhedral in all cases. Albite is formed at a temperature of 180oC or
higher (Brown, 1978), and its presence implies zones of low permeability.
Epidote was first noted at 1542 m through petrographic analysis from its high relief, pale
green to pale yellow and greenish brown pleochroism, in addition to parallel extinction. The
interference colors are distinctively bright (second order). Generally, epidote is sporadically
distributed in this well and is found more commonly as a replacement mineral of the primary
feldspars (plagioclase and sanidines) and to a lesser extent the pyroxenes. It occurs at discrete
depths and disappears completely below 2784 m. The minerals that were commonly found
in association with epidote include mixed-layer clays (MLCs), secondary quartz, calcite,
pyrite and titanite. In drill cuttings analysis, epidote was identified by its conspicous yellow
to greenish yellow colour. Pure crystalline epidote is rarely present and in most instances
(e.g. 2234-2236 m and 2570-2580 m) the mineral was observed only as a faint coloration.
In most geothermal systems, the appearence of epidote indicates temperatures equal to or
above 240oC (Bird and Spieler, 2004).
Clay minerals
The ubiquitous nature of clay minerals as a function of physico-chemical environment of
crystallization has enabled their application as markers of paleo-conditions in active
geothermal systems. Geothermal systems are especially fruitful places to study clay minerals
genesis since these environments serve as natural laboratories where fluid-rock interactions
occur under measurable conditions (Steiner, 1968). The distribution of clay minerals in
active geothermal fields depends on the ability of the fluid to approach equilibrium with host
rocks at any scale during the hydrothermal processes. Clay mineral transformations are
governed by the Ostwald Step Rule (Moore and Reynolds, 1997). According to the rule, clay
mineral paragenesis evolves through the formation of sucessive metastable phyllosilicates
35
that progress towards the state of stable chemical and textural equilibrium. The minerals
have proved to be particularly useful for temperature estimations (mineral geothermometers)
based either on modifications of crystal structure (crystallinity, interstratification, order-
disorder, etc) or modifications of crystal chemistry (layer charge, substitutions, etc) of the
crystallites (Moore and Reynolds, 1997). Clay minerals in well OW-922 occur largely as
replacement products of volcanic glass, feldspars, ferromagnesian minerals and rarely as
depositional products in veins and vesicles. In the latter cases, the minerals are found in
association mostly with hydrothermal quartz. Identification of the clay minerals was
primarily based on the position of d-spacings determined with XRD after air-drying,
saturation with ethylene glycol and heat treatment. X-ray diffraction analysis of drill cuttings
from well OW-922 revealed four distinct minerals of the phyllosilicate group as discussed
below. Representative X-ray data for clay minerals observed in this well are compiled in
Appendix B, and selected X-ray diffractograms shown in Figures 15 and 18.
Smectite
Smectite, also known as ‘‘swelling clay” is a low-temperature clay alteration mineral and
was first noted at 338 m depth through XRD analysis. It forms at comparatively low
temperatures, up to 120oC, though generally not exceeding about 200oC (Brown, 1984). In
the study well, the occurrence of smectite is widespread, occurring at shallow depths of the
well (338-450 m) and extending to depths between 1138-1166 m. The mineral is noted
largely replacing volcanic glass and ferromagnesian minerals. In binocular analysis, the
mineral was identified by its extremely fine-grained and poorly crystallized nature,
particularly in the uppermost 450 m. The color varies from light to dark green to brown.
According to Steiner (1968), smectite is commonly the first clay mineral phase to replace
volcanic glass and likely represents a metastable phase of reaction progress involving glass
and hydrothermal fluids. Diffraction patterns of the untreated air-dried samples show broad
d(001) peaks ranging from 12.4 Å to 15.0 Å (Figure 15). The basal diffraction peak is quite
sensitive to ethylene-glycol treatment and invariably widens to higher spacing, ranging from
16.9 Å to 17.5 Å. However, the peak irreversibly collapses upon heating to about 9.97 Å
(Figure 15), corresponding to the theoretical spacing for ideal smectite. The observed
changes of the first-order, basal reflection, d-spacing between peaks is caused by variation
of interlayer water molecules.
Laird (2006), observed that smectites swell by absorbing water or polar organic solvents
between smectite quasicrystals and/or between the individual layers within quasicrystals.
Such swelling occurs when the clay is dispersed in a solvent, or when the clay is in direct
contact with an atmosphere having a high vapor pressure of the solvent. Smectite is
associated with a secondary mineral assemblage consisting principally of pyrite, zeolites and
quartz. In many geothermal systems (Steiner, 1968), smectite has been noted gradually
transforming to illite in response to increasing temperature or depth through a series of
progressively more ordered smectite/illite interlayer structure. In contrast, cases of illite
converting back to smectite have equally been observed in other geothermal fields, e.g. in
Wairakei (Steiner, 1968), particularly where there is an inversion of temperature.
36
Mixed-Layered Clays (MLCs)
The term MLCs is synonymous with interlayered or interstratified clay minerals. Essentially,
MLCs represent clay minerals derived from two or more kinds of intergrown layers as
opposed to physical mixtures (Moore and Reynolds, 1997). Most often, MLCs contain
smectite as a swelling/expanding component. The interlayering is defined by three particular
types, i.e. regular (ordered), random or segregated (Moore and Reynolds, 1997). In well
OW-922, these clays mark the transitional stage from swelling to non-swelling varieties.
They are noted as the most predominant clay mineral phases, occurring regularly to the
bottom of the well. MLCs apparently can form in many possible combinations of different
layers. However, the common varieties in this case are smectite/illite, smectite/chlorite and
the three component type, smectite/illite/chlorite. The latter is the most abundant and
widespread form, occurring to the well bottom. It is characterised by a d-spacing of 15.4-
12.2 Å for untreated samples and 17.8-16.5 Å for ethylene glycol solvated samples. The
expanded layers collapse to about 9.98 Å upon heating. The chlorite component is indicated
by the 7 Å peak which collapses totally as a result of oven heating. MLCs were not identified
in binocular analysis but were recognized by the relatively more yellowish to green colour,
stronger pleochroism and brighter birefringence colours (Figure 16) relative to chlorite in
optical petrography. MLCs were first noted at 1038 m through XRD analysis (Appendix B).
EMP analysis (analyses not shown) show relatively higher K2O (7.7-9.8 wt.%) for MLCs
compared to chlorite (0.12-0.62 wt.%). These clays are normally stable at temperatures ≤
220oC.
OW-922 358-360 m
File: 358-360H_1.raw - Type: 2Th/Th locked - Start: 2.000 ° - End: 15.000 ° - Step: 0.020 ° - Step time: 1. s - Temp.: 25 °C (Room) - Time Started: 0 s - 2-Theta: 2.000 ° - Theta: 1.000 ° - Phi: 0.00 ° - Aux1: 0.0 - Aux
File: 358-360G_1.raw - Type: 2Th/Th locked - Start: 2.000 ° - End: 15.000 ° - Step: 0.020 ° - Step time: 1. s - Temp.: 25 °C (Room) - Time Started: 0 s - 2-Theta: 2.000 ° - Theta: 1.000 ° - Phi: 0.00 ° - Aux1: 0.0 - Aux
File: 358-360A_1.raw - Type: 2Th/Th locked - Start: 2.000 ° - End: 15.000 ° - Step: 0.020 ° - Step time: 1. s - Temp.: 25 °C (Room) - Time Started: 0 s - 2-Theta: 2.000 ° - Theta: 1.000 ° - Phi: 0.00 ° - Aux1: 0.0 - Aux
Lin
(C
ou
nts
)
0
100
200
300
400
500
2-Theta - Scale
2 3 4 5 6 7 8 9 10 11 12 13 14 15
d=
16,7
6951
d=
12,4
8563
d=
9,9
7703
d=
6,4
6542
d=
8,4
1491
Smectite
Figure 15: X-ray diffractogram of air dried, glycolated and heated sample of smectite.
37
Illite
Illite was first noted at 474 m through XRD analysis. The mineral occurs at temperatures of
approximately 200oC or more. It is observed to form mostly from the alteration of feldspars,
even though precipitates in veins and vesicles are also noted in small quantities. The mineral
is associated with a mixture of secondary minerals including variable but generally trace
amounts of hydrothermal quartz, chlorite, and pyrite. In thin section, illite is colourless to
yellowish brown, with the texture varying between fine to coarse. In many cases (e.g.
between 1238-1240 m), it was observed as matted flakes or fibrous aggregates. In XRD
analysis, the diffraction patterns for the untreated sample show a first order basal reflections
at ~ 10 Å and do not show significant departure from the 10 Å integral series upon ethylene
glycol solvation and heating.
Chlorite
Chlorite is a common clay mineral in the Olkaria geothermal system whose occurence
indicates reservoir temperatures exceeding 220o C (Kristmanndóttir, 1979). In well OW-922,
chlorite formed as a replacement product of ferromagnesian minerals (mostly pyroxenes)
and feldspars (sanidine and plagioclase), and also as densely packed aggregates filling
vesicles. It occurs preferentially in basalts and trachyte and is associated with mineral
assemblages of quartz, calcite and pyrite, and rarely, epidote. According to the XRD results,
pure chlorite is seldom found as an individual phase, but in many cases the mixed-layer clays
has chlorite structure layers along with smectite and illite layers (Appendix B).The colour
varies widely from green to bluish green as observed in binocular investigation. In thin
section, the mineral is fibrous and displays light to dark green colours, weakly pleochroic
from yellow green to green (Figure 17). In addition, it has low interference colours (first
order gray). Occasionally, the mineral occurs in minute amounts forming spherulites.
Figure 16: Cross polarized image of mixed-layer clay depicting high
interference colours at a depth of 2782 m.
38
In XRD analysis, the spacings of the first order d(001) basal reflections at ~12.4-14.7 Å are
not affected by air drying, glycol solvation or heat treatment. This peak however, is obscured
in most samples in this well (e.g. Figure 18), leaving the second order reflections d(002)
stable at ~7.0-7.1 Å for untreated and glycol treatment as the only conspicuous peaks. The
latter, totally disappears upon oven heating though. Previous studies (e.g. Otieno and Kubai,
2013) have termed this chlorite as an unstable variety and it is the most common chlorite
type in many wells in the Olkaria sub-fields. A widely quoted hypothesis by Moore and
Reynolds (1997) suggests this kind of chlorite to be Fe-rich. EMP analysis results for
chlorite, discussed in section 4.9, are fully in agreement with this observation. Based on their
argument, the Fe-rich component in the chlorite is introduced by prolonged interaction
between the adjacent host rocks and the altering solution.
Figure 17: A plane polarized image of fibrous chlorite filling a vesicle
at 1826 m.
39
4.2.2 Hydrothermal alteration zonation
Many high-temperature geothermal systems of the world are, divisible into different zones
based on changes in alteration assemblages. Different alteration minerals become stable or
metastable within a particular temperature range and with certain fluid compositions. Thus,
by mapping the mineral sequences as well as the alteration minerals, it is possible to deduce
the thermal evolution and conditions of a specific geothermal system (e.g. Mortensen et al.,
2014). Alteration zones are characterised by an abundance of particular minerals
corresponding to increased temperature and depth. The top of each alteration zone is defined
by the depth of the first appearance of a mineral. Four alteration zones, namely zeolite-
smectite, illite-chlorite, mixed-layer clay zone and epidote-mixed-layer clay zone, were
distinguished below the unaltered zone based on optical petrography and binocular
observations. Representative alteration mineral assemblages of each zone are listed as well.
Identification of the different clay types in this well relied largely on XRD results (Figure
14), and to a lesser extent on electron microprobe analysis.
0-190 m depth: Unaltered zone: This is the lowest grade alteration zone in this well. It is
composed of relatively fresh rocks, mainly pyroclastics with no alteration signature related
to geothermal activity, an indication that the temperatures have been < 50oC. Volcanic glass
constitutes the bulk of the rocks and are conspicuous both in binocular and petrographic
observation. The unaltered zone is characterised primarily by oxidation, which supposedly
OW-922 1860-1862 m
File: 1860-1862H_1.raw - Type: 2Th/Th locked - Start: 2.000 ° - End: 15.000 ° - Step: 0.020 ° - Step time: 1. s - Temp.: 25 °C (Room) - Time Started: 0 s - 2-Theta: 2.000 ° - Theta: 1.000 ° - Phi: 0.00 ° - Aux1: 0.0 - A
File: 1860-1862G_1.raw - Type: 2Th/Th locked - Start: 2.000 ° - End: 15.000 ° - Step: 0.020 ° - Step time: 1. s - Temp.: 25 °C (Room) - Time Started: 0 s - 2-Theta: 2.000 ° - Theta: 1.000 ° - Phi: 0.00 ° - Aux1: 0.0 - A
File: 1860-1862A_1.raw - Type: 2Th/Th locked - Start: 2.000 ° - End: 15.000 ° - Step: 0.020 ° - Step time: 1. s - Temp.: 25 °C (Room) - Time Started: 0 s - 2-Theta: 2.000 ° - Theta: 1.000 ° - Phi: 0.00 ° - Aux1: 0.0 - A
Lin
(C
ou
nts
)
0
100
200
300
400
500
600
700
800
2-Theta - Scale
2.1 3 4 5 6 7 8 9 10 11 12 13 14 15
d=
7,1
5824
d=
6,3
9214
d=
10,0
1907
d=
14,1
8330
Illite and Unst. chlorite
Figure 18: X-ray diffractogram of illite and unstable chlorite showing obscured
peak of 14.18 Å for the three treatments and 7.15 Å peak for air dried and glycol
solvation. The latter peak totally collapses upon heating.
40
is related to the interaction between the host rocks and the comparatively oxygen-rich
descending cold groundwater. Total loss of circulation between 96-190 m (Appendix A)
made it impossible to locate precisely where this zone terminates within the specified depth
range.
190-474 m depth: Zeolite-smectite zone. The alteration mineral assemblages in this zone are
characterised by low-temperature zeolites, smectite clay and trace quantities of pyrite and
calcite. Other notable minerals include chalcedony and titanite. The low-temperature zeolites
present in this zone are mesolite and scolecite. Smectite was first recognized by XRD
analysis at 338 m (Appendix B), suggesting that the temperatures in this zone are < 180-
200oC.
474-1038 m depth: Illite-chlorite zone. The concurrent appearence of illite and chlorite at
474 m defines the top boundary of this zone, signifying that the formation temperature is
gradually increasing with depth and is in excess of 200oC. In conjunction with this, the
zeolites (mesolite and scolecite) show indications of instability and completely disappear at
482 m. Quartz and wairakite appear and are first identified at 476 m and 494m, respectively.
According to XRD analysis, the chlorite is an unstable variety. Other alteration minerals
found in this zone are chalcedony, titanite, calcite, albite and pyrite. Calcite is ever-present
as open-space and matrix fillings, while pyrite is prevalent as patchy disseminations in the
rock groundmass.
1038-1542 m depth: Mixed layer clay (MLC) zone. The characteristic feature of this zone is
the first appearence of MLCs, in the form of smectite/chlorite at 1038 m. Based on the
discussion under MLCs, temperatures in this zone are ≤ 220oC. Common assemblages
continue to include pyrite, calcite, haematite and titanite.
1542-2990 m depth: Epidote-Mixed layer clay zone. Epidote first occurs at 1542 m and
defines the top of this zone. It is the largest zone extending to the well bottom. The first
appearence of epidote indicates that the formation temperature has been at or higher than
240oC. The occurence of this mineral is very sparse in comparison to MLCs. Epidote is
observed to occur until about 2784 m, after which it totally disappears to the bottom of the
well. In addition to MLCs, epidote, quartz, titanite, chalcedony, haematite and pyrite, calcite
is by far the dominant mineral phase. The mineral is observed overprinting epidote at 2782
m.
4.2.3 Distribution of alteration mineral zones in the Domes area
Alteration mineral zones are basically the result of diagnostic alteration minerals present
within a particular geothermal system, and can give reliable indications of the reservoir
temperature. The formation of hydrothermal minerals, as discussed above is not solely a
function of temperature, but also fluid composition, initial composition of the rock system,
permeability, etc (Brown, 1978). In Figure 19, the distribution of alteration mineral zones in
the study well is compared with results of previous studies (e.g. Musonye, 2015; Mwangi,
2012). From this figure, it is evident that zones indicative of high temperatures, e.g. illite-
chlorite-epidote and chlorite-epidote-actinolite, marked by the first appearence of epidote
and actinolite, respectively are present at relatively shallow depths ( between 1280-1440
masl and 800-1240 masl, respectively) in wells OW-910, 916 and 914. The same zones
appear at relatively greater depths in well OW-917. However, in well OW-922, the two zones
are conspicuously absent, even though it is known from hydrothermal mineralisation that the
41
first appearence of epidote in this well occur at even much deeper depth, i.e. at 458 masl.
Observations made for the distribution of alteration mineral zones across a section of the
Domes field positively correlate with the model for measured formation temperature
distribution across the same field (Figure 35). High formation temperatures are observed at
comparatively shallow depths in wells OW-910, 916 and 914, whereas, at greater depths in
well OW-917. In contrast, well OW- 922 is generally characterised by low temperatures.
Figure 19: Distribution of alteration mineral zones across a
section of the Domes area, along the traverse A-B.
42
4.3 Feed zones
Feed zones are basically water-saturated regions in the sub-surface which produce an
economically feasible quantity of water to a well. The primary goal of drilling a geothermal
well is to intersect as many permeable and hot feed zones as possible, below the production
casing, so as to maximise the well productivity. For locating feed points during drilling, one
can rely on a combination of observations, including, circulation losses or gains, intrusive
boundaries, lithological contacts, changes in alteration mineralogy patterns, appearance of
highly fractured formations, changes in circulating fluid temperatures, as well as temperature
recovery logs. In the Domes area, the location of feed points from previous studies (e.g.
Mwangi, 2012; Ronoh, 2012; Lagat, 2004) have been inferred primarily from lithological
boundaries, highly fractured lavas, sudden increase in the rate of penetration and variation
in alteration patterns.
In drill cuttings analysis of the study well, fluid “feeder” zones below the production casing
were deduced from lithological interfaces, alteration intensities and drilling parameters (i.e.
rate of penetration (ROP)). Observations of circulation losses have equally played some role,
but not to a large extent since most of the loss zones encountered range between 2 and 14 m,
and in most cases are principally attributed to a change in stand-pipe pressure during POOH
to change a drill bit. Temperature and pressure recovery were determined by four heat-up
profiles, i.e. after 9 hours and, 12, 21 and 49 days. Pre-injection temperature and pressure
profiles were taken prior to the recovery tests. However, it is worth noting that no injection
tests were conducted since the well got filled up with water during injection, which argues
against the existence of any major feed zone(s). As a consequence, no injectivity index was
measured for this well. Another key point to highlight is that the temperature recovery over
a period of only 49 days is not sufficient for the geothermal system near the well to reach
stable conditions following the long drilling time (Figure 13).
Step pumping was conducted between 19th September and 6th November, 2014. The first
pumping rate was set at 1000 l/min, and then subsequently increased to 1300 l/min, 1600
l/min and finally 1900 l/min. Traditionally, this is the pumping rate, which has been adopted
during step pumping of the Olkaria wells. During all pumping rates, temperature and
pressure measurements were carried out at 100 m interval from the surface to 1200 m, which
is the production casing depth. Afterwards, the interval was reduced to 50 m to the well
bottom. These intervals are constant for all wells. However, the change from 100 to 50 m is
subject to the casing depth, which is quite variable from one well to another. Pressure logs
measured in this well (Figure 20) display no intersection of the various pressure recovery
profiles, commonly referred to as the "pivot point".
43
From the temperature recovery logs (Figure 21), it is observed that temperature increases
from the surface to about 2250 m at an average rate of 4.5o C/100 m depth. This is followed
by a more or less constant temperature between 2300-2500 m depth. Below 2500 m, a steady
decline (~ 3o C/100 m) is noted in all the five temperature logs to the bottom of the well. The
increase in temperature from the surface to about 2250 m depth coincides with the trend of
increasing alteration grade. A prograde change in alteration grade, cutting across different
alteration zones, is observed to a depth of about 1798 m. This is consistent with the transition
from the low-temperature alteration minerals, zeolites, smectites and chalcedony to
relatively high-temperature minerals, such as quartz, albite, wairakite, illite, chlorite and
epidote (Figure 14).
Five small feed zones have been recognised from the study well. The classification criterion
is based on the relative size of kink in temperature recovery logs and the extent of geological
observations. These are located below the production casing at depths between 1260-1312
m, 1840-1900 m, 2070-2162 m, 2210-2276 m and 2360-2444 m. The feed zones are clearly
evident from the kinks in temperature logs (Figure 21), and are supported by both drilling
and geological observations. Feed zones noted above the production section (e.g. between
96-190 m) are of no consequence since this region has been completely cased off. However,
such feed zones are costly to the drilling operations because they require large tonnes of
cement to seal. Therefore, it is necessary to map and note their geological extent as a
precaution to subsequent wells planned to be drilled on the same well pad. The feed zone
between 1260-1312 m is hosted within a highly vesicular and altered tuff, marked by
extensive pyritization. A remarkably high rate of penetration (4.8-10.2 m/hr) is observed
within this zone, lending support to the geological observations.
Figure 20: Downhole pressure recovery logs for well
OW-922.
44
Feed zone between 1840-1900 m, is hosted within a highly altered and extensive trachytic
formation (Figure 14), characterised by numerous loss zones. Also, a deflection both in the
temperature profile and ROP profile is noted within this zone. Feed zones between 2070-
2162 m and 2210-2276 m are located within highly vesicular and moderately to highly
altered trachyte, characterised by several minor (2-10 m) loss zones. The final feed zone
between 2360-2444 m is located at the lithological interface of trachydacite and rhyolite. It
is also characterised by moderately to highly altered rhyolite, intense calcite deposition and
several loss zones, ranging in size between 2 and 14 m. Temperature recovery profile
indicates the highest formation temperature (127oC) measured in the well, occur within this
interval, despite the zone being associated with scarcity of high temperature alteration
minerals. The observed feed zones are summarized in Table 4.
45
Figure 21: Feed zones in well OW-922 as deduced from temperature recovery logs,
drilling parameters and geological observations.
46
Table 4: Interpreted permeable zones based on geological observations.
Depth (m) Evidence based on geological observations Size
1260-1312 Highly altered and vesicular tuff
Prevalence of pyrite deposition
Small
1840-1900 Highly altered trachyte
Numerous circulation losses
Small
2070-2162 Highly altered and vesicular trachyte
Numerous circulation losses
Small
2210-2276 Highly altered and vesicular trachyte
Numerous circulation losses
Small
2360-2444 Lithological interfaces between trachydacite
and rhyolite.
Moderate to high alteration intensity
Several circulation losses
Intense calcite deposition
Small
4.4 Alteration and measured formation
temperatures
Studies of temperature dependent alteration minerals, temperature logging (direct down-hole
measurements) and fluid inclusion homogenization temperatures (Th) have successfully
been applied to assess the physico-chemical conditions and temporal evolution of
geothermal reservoirs in active geothermal fields (e.g. Marks et al., 2010). A hydrothermal
alteration sequence provides significant insights into the history of a geothermal system.
There is always some uncertainty whether the observed alteration mineral assemblages
reflect the present formation temperatures or are related to some past thermal events. Like
in many geothermal systems, hydrothermal alteration minerals in the Olkaria geothermal
field have successfully served as important indicators of sub-surface thermal changes (e.g.
Lagat, 2004; Mwangi, 2012; Okoo, 2013).
A correlation of the three methods leads to reconstruction of a more plausible and realistic
model of a geothermal system. Fluid inclusion measurements were, however, not conducted
on samples from the study well. As a result, reconstruction of the thermal history of the
geothermal system around well OW-922 is based exclusively on inferred alteration and
measured formation temperatures (Figure 22). Alteration temperature is used to present
temperature in a geothermal system, and in many cases is assumed to represent the maximum
long-term temperature state. In contrast, measured formation temperature represents the
current state of the system. Alteration mineral curve plotted in Figure 22 relied on the first
appearance of alteration minerals observed, namely quartz, wairakite and epidote. The three
minerals have their first appearence at depths of 476 m, 494 m and 1542 m, respectively.
Temperature profile was plotted based on a 49 day heat-up period, i.e. the last temperature
profile measured in the well.
47
4.5 Whole-rock chemistry
Bulk-rock chemical analysis is important in understanding the chemical diversity of the
samples from the study well and the conditions under which the magma was generated. To
achieve this, plots have been made, including TAS diagram, Al2O3 vs. FeO and several
Harkers' variation diagrams based on the chemical analytical results. Generally, any sample
has two types of elements: major and trace (including rare earth elements).
4.6 Major and trace element Compositions
Thirty representative samples were analysed for whole-rock chemistry (major and trace
elements) in the study well. Concentrations of 10 major element oxides and 13 trace elements
were determined using the ICP-OES procedure outlined in Appendix C.
Figure 22: Comparison of inferred alteration and measured
formation temperatures along with boiling point curve for well OW-
922.
48
4.6.1 Major elements
Rocks in well OW-922 exhibit significant variations in major element compositions. Mafic
and intermediate rock suites have striking similarities, depicting a homogeneous trend for
FeO, MgO, and CaO, all showing a continuous decrease with increase in SiO2 content for
most of the samples (Figure 23). FeO shows a limited range (10.2-10.6 wt.%) in the mafic
and intermediate rocks and, slightly low abundance (3-8 wt.%) in the highly evolved rocks.
MgO is highly depleted in the highly evolved rocks (0.03-0.7 wt.%), whereas, the content
ranges between 3-6 wt.% in the mafic and intermediate rocks. CaO has a more or less similar
pattern to FeO, depicting a restricted range of between 6-11 wt.% in the mafic and
intermediate rocks and declining significantly to an average of about 1 wt.% in the most
silicic rocks.
TiO2 exhibits a limited range in the mafics and intermediate rocks (1.83-2.08 wt.%) and
decreases rapidly with increasing SiO2 content (Table 5). P2O5 displays considerable scatter
and no obvious variation trend, even though its contents are observed to decrease with
increase in SiO2 concentration. The scatter is more pronounced in the basic and intermediate
fields. Like TiO2, P2O5 is marked by a narrow range in the mafic and intermediate rocks
(0.18-0.60 wt.%). The contents decrease from about 0.60 wt.% in the mafic rocks to just
under 0.04 wt.% in the more silicic rocks, that is at approximately 69-73 wt.% SiO2. On the
other hand, MnO shows a nearly flat trend without any systematic variation with SiO2
(Figure 23). The elements' contents are fairly constant between 0.13 and 0.28 wt.%. As a
result, little emphasis is given on this element in interpretation of the data.
Al2O3 contents range from 10.3-17.04 wt.% in all the rock types. The element initially
exhibits a more or less similar picture as MnO, but displays modest decline with increasing
SiO2 contents for most of the samples. Na2O and K2O correlate positively with SiO2 at the
initial stage. The two oxides have a very similar pattern even though sodium increases
relatively more than potassium. The concentration of the two oxides increases and attains a
peak at approximately 68 and 64 wt.% SiO2 for potassium and sodium, respectively,
followed by a slight decrease with increasing SiO2 contents. The decrease is more prominent
in sodium, particularly for the samples at the greatest depths (i.e. between 2420 and 2990 m)
in the well. In contrast, K2O content appears to flatten out with a further increase in SiO2
concentration. The two elements are generally lower in the mafics and intermediate rocks
compared to the evolved rocks. In particular, Na2O concentration is distinctly higher than
K2O, with the exception of a few samples in the trachytes and trachydacites. The composition
of the two elements in rhyolites is, however, quite variable. Peralkalinity index (i.e. P.I. =
mol. [Na+K]/Al) of the samples analysed ranges between 0.26 and 1.03.
49
1
2
3
4
5
6
7
8
45 50 55 60 65 70 75
Na 2
O (
wt.
%)
6
8
10
12
14
16
18
Al 2
O3
(wt.
%)
0
2
4
6
8
45 50 55 60 65 70 75
K2O
(w
t.%
)
0,0
0,1
0,2
0,3
45 50 55 60 65 70 75
Mn
O (
wt.
%)
0,0
0,5
1,0
1,5
2,0
2,5
45 50 55 60 65 70 75
TiO
2(w
t.%
)
0
2
4
6
8
10
12
14
45 50 55 60 65 70 75
CaO
(w
t.%
)
0
2
4
6
8
45 50 55 60 65 70 75
MgO
(w
t.%
)
SiO2 (wt.%)
0
2
4
6
8
10
12
45 50 55 60 65 70 75
FeO
(w
t.%
)
SiO2 (wt.%)
Figure 23: Harker diagrams showing major element variations against SiO2 for samples
from well OW-922.
50
Figure 23 (continued): Harker diagrams showing major element oxides and CaO/ALK
variations agains SiO2 for samples from well OW-922.
4.6.2 Trace elements
The term trace element lacks a rigid definition, even though several authors (e.g.
Harðarson, 1993) have used this term in many instances to refer to those elements present
in rocks in concentrations of a few parts per million (ppm). The significant features of the
trace element concentrations observed from the variation diagrams (Figure 24) as
differentiation progresses from the mafic to silica rich rocks are: (1) the progressive
depletion in the concentrations of the transition metals, Sc, Co, V, Cr, Ni, and, Sr with
increase in SiO2 contents, (2) Incompatible trace elements (ITEs), such as the HFSE, Y and
Zr; Rb and LREE, La, show significant enrichment with increase in SiO2 contents.
Vanadium decreases approximately from 300 ppm to 168 ppm from the mafic to
intermediate rocks. Subsequently, an abrupt drop in V content down to 0.5 ppm is observed
in the highly evolved rocks. Cr content is high in the mafic rocks (132 ppm) but exhibit a
remarkable decline in the intermediate rocks (17-59 ppm). The concentration further
decreases to between 2-27 ppm in the silicic rocks. Sc vs. SiO2 plot shows a small but
consistent decrease in Sc content between the mafic and intermediate suites (47-25 ppm)
until about 69 wt.% SiO2 where the drop is rather steep (0.2-0.3 ppm). The rhyolitic rocks
are more depleted in Sc compared to the other evolved rock suites. A fair amount of scatter
is apparent in Sr concentration in the mafic and intermediate rocks as well as in two
trachytic samples (1878 m and 1922 m). The element shows a decrease from 376 ppm to
256 ppm as SiO2 increases to about 55 wt.%. A significant decrease to about 8 ppm,
thereafter follows in the more evolved derivatives. Two trachytic samples at 1878 m and
1922 m depth, however, contain anomalous Sr contents, 307 ppm and 344 ppm at relatively
high SiO2 contents, i.e. 58 wt.% and 63 wt.%, respectively.
Rubidium shows high abundance (123-226 ppm) with increase in SiO2 (64-73 wt.%). A
significant decline (≤ 78 ppm) is observed in the mafic and intermediate rocks,
characterised by SiO2 range of between 49-55 wt.%. Y displays a very similar trend to Rb,
showing a gradual increase with increasing SiO2 concentration. In particular, it is more
enriched (101-304 ppm) between 67-73 wt.% SiO2. La follows similar pattern as Rb and
Y. The element shows a concentration of between 110-424 ppm at SiO2 content of between
63-73 wt.%. The contents, however, fall to < 80 ppm in the mafic and intermediate rocks.
0,0
0,2
0,4
0,6
0,8
45 50 55 60 65 70 75
P2O
5(w
t.%
)
SiO2 (wt.%)
0,0
0,5
1,0
1,5
2,0
2,5
3,0
3,5
45 50 55 60 65 70 75
CaO
/ALK
SiO2 (wt.%)
51
Zr shows significant increase with increasing SiO2 content but scatters mostly between
2793 and 2995 ppm. Ba, on the other hand exhibits wide scattering in whole-rock
compositions with no well defined trend. The element occurs in considerable abundances
in the intermediate rocks, even though exceptionally high abundances (1775-2238 ppm)
are observed in the trachytes. However, Ba is strongly depleted in the rhyolites, in most
cases falling below 20 ppm. Results of major and trace element concentrations are
presented in Table 5.
0
100
200
300
400
45 50 55 60 65 70 75
Sr (
pp
m)
0
10
20
30
40
50
45 50 55 60 65 70 75
Sc (
pp
m)
0
100
200
300
400
45 50 55 60 65 70 75
V (
pp
m)
0
40
80
120
160
45 50 55 60 65 70 75
Cr
(pp
m)
0
20
40
60
80
100
120
45 50 55 60 65 70 75
Ni (
pp
m)
SiO2 (wt.%)
0
10
20
30
40
50
60
45 50 55 60 65 70 75
Co
(p
pm
)
SiO2 (wt.%)
Figure 24: Variation diagrams for a range of trace elements plotted against the
index of differentiation, SiO2 for samples from well OW-922.
52
4.7 Classification of rock types
Petrologists have devised several schemes for the chemical classification of non-pyroclastic
volcanic rocks. In spite of that, none has been found fully satisfactory. Many systems were
in use before classification based on the amount of alkali (Na2O + K2O) versus silica (SiO2)
diagram in igneous rocks, popularly known as the “TAS diagram” (Figure 25), was
eventually accepted by the International Union of Geological Sciences (IUGS). The
variation of silica and total alkalis is a highly useful discrimination technique. The total alkali
0
100
200
300
400
45 50 55 60 65 70 75
Y (p
pm
)
0
100
200
300
400
500
45 50 55 60 65 70 75
La (
pp
m)
0
500
1000
1500
2000
2500
45 50 55 60 65 70 75
Ba
(pp
m)
0
50
100
150
200
250
45 50 55 60 65 70 75
Rb
(p
pm
)
0
500
1000
1500
2000
2500
3000
3500
45 50 55 60 65 70 75
Zr (
pp
m)
SiO2 (wt.%)
0
40
80
120
160
200
45 50 55 60 65 70 75
Cu
(p
pm
)
SiO2 (wt.%)
Figure 24 (continued) : Variation diagrams for a range of trace elements plotted against
the index of differentiation, SiO2 for samples from well OW-922.
53
axis essentially distinguishes alkaline from sub-alkaline rocks types, whereas the silica axis
distinguishes between evolved and primitive rock types. An important contributing factor
that led to the acceptance of this method was its simplicity, which is an essential aspect of a
classification scheme for volcanic rocks (Le Bas et al., 1986).
Generally, the compositional variation seen on a TAS diagram may be regarded as a result
of two basic processes, namely partial melting and fractional crystallisation, even though
magma mixing can be important in some cases. In this respect, decreasing degree of partial
melting results in increased total alkalis and lower SiO2. Fractional crystallisation, on the
other hand brings the rocks to both higher alkali and silica content (e.g. Harðarson, 1993).
In this section, petrochemical results from the study well are presented and compared with
the results of the previously studied wells (Musonye, 2015) aimed at understanding the
chemical composition and petrogenetic mechanisms responsible for the evolution of OVC
sub-surface rocks. Additionally, the results are compared with the findings for surface
samples for Olkaria and the minor eruptive centres surrounding the complex.
4.7.1 TAS classification scheme
In the TAS classification scheme, the concentrations of the major elements have been re-
calculated to 100% on a volatile-free basis to ensure comparability and are plotted as such
in the variation diagrams. From the classification of the rocks in the TAS system shown in
Figure 25, it is apparent that rocks from well OW-922 show a range of compositions. The
sub-alkaline rocks vary from basalt through trachydacite to rhyolite, whereas, the alkaline
rocks range from basaltic trachyandesite through tracyandesite to trachyte.
Figure 25: TAS diagram showing the compositional range for sub-surface rocks from
well OW-922, according to Le Bas et al. (1986).
54
The prevalence (~ 85-90%) of evolved rocks is noticeable, depicting a relatively wide range
in SiO2 (59-73 wt.%). Basic and intermediate rocks constitute about 10% of the intersected
succession, and have a SiO2 range of between 49.7-55 wt.% (Table 5). A large volume of
the salic differentiates cluster to a great extent in the trachydacite, rhyolite and trachyte
fields, with the latter being the dominant rock type. A few samples are, however, transitional
in composition, falling along the boundary separating the alkaline from sub-alkaline series.
Based on petrographic analysis, the rocks from well OW-922 are variably porphyritic with
plagioclase and clinopyroxene as the dominant phenocrysts phases, together with olivine and
opaques (Fe-Ti oxides) in the mafic to intermediate samples. In the more salic varieties, K-
feldspar (sanidine) is the predominant phenocryst phase, although plagioclase and
clinopyroxene (aegerine-augite) are also noted in considerable amounts.
4.7.2 Al2O3 vs. FeO classification scheme
This is a classification scheme put forward by Macdonald (1974), using the correlation
between alumina and total iron (as FeO) plot to evaluate the relationship of the samples
based on the degree of peralkalinity. Peralkaline rocks are basically divided into comendites,
pantellerites, comenditic trachytes and pantelleritic trachytes (Macdonald and Scaillet,
2006). By definition, comenditic rocks refers to the mildly peralkaline varieties, whereas,
the pantelleritic counterparts, refers to more strongly peralkaline and more Fe-rich rocks
(FeO>Al2O3) (Macdonald, 1974). None of the analytical samples fall under the definition of
pantelleritic, as in all cases, Al2O3 is observed to be greater than FeO (Table 5).
Based on this classification scheme (Figure 26), it is evident that rocks from the study well
traverse the division between comenditic trachyte, pantelleritic trachyte and pantellerite,
with the last one being the dominant variety. No samples plot within the comendite field.
Most of the trachytes plot in the comenditic trachyte field, with a few exceptions plotting as
pantelleritic trachyte and pantellerites. Rhyolites, on the other hand plot within the
pantellerites field. From the Al2O3 vs. FeO plot, it is clear that the differentiation trends for
these samples are appreciably variable.
55
Figure 26: Al2O3 vs. FeO classification (modified from Macdonald, 1974) of sub-
surface rocks from well OW-922.
56
Table 5: Major and trace element results for well OW-922 based on ICP-OES analysis.
Major elements (wt.%)
Depth 864 m 966 m 1106 m 1142 m 1208 m 1332 m 1672 m 1704 m 1734 m 1808 m 1878 m 1922 m 1966 m 2056 m 2088 m
SiO2 61.85 63.57 69.17 69.36 49.67 62.74 54.62 65.85 64.65 64.89 63.59 58.99 64.71 65.26 65.55
Al2O3 14.50 13.84 10.50 10.30 14.94 15.34 15.10 13.70 15.30 14.40 17.04 14.76 14.02 12.96 13.32
FeO 8.08 8.29 8.11 8.05 10.66 7.06 10.21 7.65 6.34 6.26 3.31 8.88 7.32 6.87 7.18
MnO 0.27 0.28 0.24 0.24 0.21 0.25 0.21 0.23 0.21 0.20 0.13 0.21 0.27 0.24 0.25
MgO 0.25 0.03 0.05 0.06 6.84 0.35 3.38 0.42 0.37 0.33 0.56 0.68 0.51 0.59 0.22
CaO 1.96 1.31 0.57 0.53 11.67 1.17 6.64 1.44 1.10 1.14 2.02 2.58 1.02 1.47 0.54
Na2O 6.83 6.95 5.90 6.00 2.98 6.73 4.44 4.14 5.69 6.25 7.04 6.35 5.13 6.23 5.81
K2O 5.28 4.93 4.61 4.57 0.87 5.38 2.47 5.47 5.14 4.97 4.93 4.85 5.88 5.17 5.95
TiO2 0.75 0.64 0.42 0.44 1.83 0.62 2.08 0.82 0.93 1.04 0.91 1.67 0.88 0.93 0.92
P2O5 0.10 0.02 0.01 0.01 0.18 0.08 0.60 0.08 0.13 0.25 0.20 0.72 0.09 0.10 0.11
Total 99.87 99.87 99.58 99.56 99.85 99.72 99.76 99.79 99.85 99.72 99.73 99.68 99.82 99.82 99.85
Trace elements (ppm)
Ba 482.52 26.48 0.00 0.00 255.88 24.22 1094.92 41.20 55.68 1775.59 1777.61 2238.54 42.70 45.74 131.01
Co 5.05 3.15 1.45 1.72 52.93 4.63 38.92 5.91 6.59 7.30 7.39 16.15 5.91 7.23 6.38
Cr 4.77 3.85 15.75 26.69 131.78 2.33 17.39 6.16 2.72 5.41 5.77 6.20 4.87 4.21 3.80
Cu 14.41 16.34 20.83 22.69 110.58 25.71 35.54 24.57 16.78 16.81 13.50 14.04 18.61 20.50 16.80
La 82.72 139.07 424.45 422.93 19.76 317.50 75.51 231.23 158.38 70.42 65.98 48.92 146.52 159.60 110.13
Ni 0.00 0.00 64.13 91.44 54.37 0.00 1.38 0.00 0.00 5.32 1.35 4.81 1.92 1.85 1.00
Rb 97.11 93.39 150.54 132.69 20.08 118.30 78.42 139.10 98.73 152.38 73.15 100.18 170.05 128.52 169.74
Sc 3.75 1.09 0.20 0.30 47.57 1.99 25.40 3.43 4.98 17.24 3.41 16.81 4.73 4.87 5.67
Sr 24.23 8.66 13.94 18.36 303.93 15.71 376.84 18.81 14.28 42.69 344.45 307.00 23.05 24.74 17.10
V 0.56 2.79 19.56 20.94 300.20 14.13 168.63 9.16 5.71 3.85 14.98 5.58 5.64 6.15 1.93
Y 51.47 92.92 302.93 282.72 26.18 193.71 49.68 145.25 100.40 71.72 34.64 55.36 133.06 124.64 101.08
57
Zn 136.74 179.65 358.59 368.48 97.19 222.95 126.65 224.30 165.60 126.90 65.26 134.52 193.55 199.89 175.22
Zr 368.19 706.84 2793.51 2984.86 100.40 1895.18 349.16 1230.88 863.51 541.83 340.04 283.14 1080.77 1102.87 773.63
Total 1271.53 1274.23 4165.89 4373.81 1520.84 2836.36 2438.43 2080.02 1493.34 2837.47 2747.52 3231.27 1831.38 1830.80 1513.49
Table 5 (continued): Major and trace element results for well OW-922 based on ICP-OES analysis.
Major elements (wt.%)
Depth 2190 m 2234 m 2288 m 2300 m 2348 m 2360 m 2420 m 2482 m 2522 m 2570 m 2600 m 2668 m 2782 m 2802 m 2982 m
SiO2 66.32 64.66 67.21 67.99 67.02 68.11 73.26 71.19 70.38 69.71 67.80 67.79 55.14 65.98 64.33
Al2O3 14.10 13.81 12.72 12.39 14.43 12.20 10.92 11.59 12.21 11.44 13.11 13.24 11.92 14.67 14.99
FeO 6.34 6.92 7.32 6.50 6.58 8.37 5.97 7.15 6.74 6.97 7.05 6.82 10.43 5.96 6.54
MnO 0.24 0.26 0.27 0.22 0.20 0.27 0.19 0.23 0.24 0.25 0.27 0.25 0.25 0.23 0.23
MgO 0.47 0.52 0.38 0.35 0.47 0.39 0.13 0.17 0.22 0.28 0.21 0.32 5.03 0.33 0.69
CaO 1.36 0.98 0.74 0.78 2.06 0.99 1.02 0.71 0.66 0.99 0.88 0.63 6.73 0.99 1.58
Na2O 6.31 5.53 4.05 4.72 4.06 5.61 1.47 3.25 3.66 2.94 4.53 4.33 4.59 5.31 5.27
K2O 3.50 5.97 6.24 5.98 3.78 3.13 6.12 4.83 4.93 6.51 5.20 5.50 3.81 5.28 4.98
TiO2 1.06 1.05 0.78 0.79 1.09 0.63 0.62 0.62 0.69 0.65 0.70 0.84 1.62 0.95 1.03
P2O5 0.14 0.15 0.08 0.06 0.11 0.04 0.04 0.01 0.04 0.03 0.05 0.07 0.19 0.13 0.15
Total 99.84 99.84 99.78 99.78 99.79 99.75 99.73 99.74 99.76 99.77 99.80 99.79 99.71 99.82 99.79
Trace elements (ppm)
Ba 149.58 127.92 64.01 58.13 325.68 100.83 20.24 12.19 10.05 18.94 13.73 69.62 214.55 229.35 299.50
Co 8.01 7.94 5.05 4.97 12.48 4.78 2.58 3.22 3.64 3.36 3.93 5.43 38.67 7.32 10.41
Cr 4.16 4.99 3.45 12.34 20.22 5.93 3.90 3.37 3.52 2.88 5.28 4.05 59.80 4.49 5.08
Cu 17.22 26.86 22.41 15.38 19.64 25.67 16.05 15.08 15.86 21.31 15.04 16.59 161.25 30.68 25.51
La 121.69 110.69 162.27 172.77 162.11 226.33 261.40 260.69 228.78 198.92 185.74 169.70 181.88 144.24 155.43
Ni 2.00 5.38 1.15 7.55 9.71 2.54 2.35 1.10 3.92 2.90 2.74 2.11 112.51 34.59 11.70
Rb 134.79 173.93 182.65 189.47 147.49 176.99 183.63 171.98 226.73 135.79 133.92 123.74 76.88 144.93 173.04
58
Sc 5.71 6.28 4.30 3.92 4.86 1.58 2.34 2.38 3.27 3.38 3.95 3.92 25.14 5.53 6.23
Sr 40.72 21.41 22.16 20.88 77.13 32.11 20.81 14.97 14.72 18.34 13.76 22.37 256.51 23.65 76.20
V 5.10 4.23 8.24 7.46 31.38 14.51 10.84 10.79 7.96 8.21 7.25 9.06 200.34 5.18 21.44
Y 100.80 99.82 140.46 137.88 124.71 178.46 176.06 185.15 151.88 139.13 131.22 141.02 125.19 95.25 106.83
Zn 148.91 175.62 230.49 199.54 201.56 283.65 210.56 227.88 214.70 245.66 215.22 203.28 248.31 162.09 153.31
Zr 837.44 810.00 1355.26 1357.33 943.30 1487.53 1807.54 1733.52 1479.58 1479.69 1275.09 1365.92 1189.73 934.39 1027.02
Total 1576.14 1575.07 2201.89 2187.62 2080.27 2540.91 2718.29 2642.31 2364.62 2278.52 2006.88 2136.81 2890.78 1821.69 2071.70
59
4.8 Comparison of well OW-922 with surface
and other sub-surface samples
This section presents a comparison of the geochemical results from well OW-922 with
surface samples from Olkaria and other minor eruption centres located along the periphery
of the complex (Figure 8). Surface data from Olkaria, together with the minor eruption
centres were acquired from previous studies of Macdonald et al. (2008), Marshall et al.
(2009) and Clarke et al. (1990). Majority of the surface samples are considered quite fresh
and apparently devoid of alteration. Additional sub-surface whole-rock chemistry data have
been obtained from Olkaria, published by Musonye (2015).
4.8.1 Comparison with surface samples
From Figure 27, it is immediately evident that samples from wells OW-922, 905A, 910 and
917 plot in more or less similar fields to the surface samples from Olkaria and neighboring
minor eruption centres. The Lolonito, Ndabibi and Akira basalts (Figure 7) plot in a similar
field to basalt and basaltic trachyandesite from the study well and the other three wells.
Equivalently, the Gorge Farm and Broad Acres rhyolites, Olkaria comendites (Figure 8) and
Tandamara trachytes (Figure 7), plot mostly in a similar position as the trachyte and rhyolites
from well OW-922.
LB NB AB
OC TT GF BA
OC- Olkaria Comendite TT- Tandamara Trachyte GF- Gorge Farm BA- Broad Acres LB- Lolonito Basalt NB- Ndabibi Basalt AB- Akira Basalt
Figure 27: TAS classification scheme showing comparison of sub-surface samples from
wells OW-922, 905A, 910 and 917 and, surface samples from Olkaria and neighboring
minor eruption centres (represented by broken lines).
60
4.8.2 Comparison with other sub-surface samples
Using selected major element oxides and trace element variation diagrams (Figures 28 and
29), a comparison has been made to check for any connection in the evolution mechanism(s)
between wells 905A, 917, 910 and 922. Once again, SiO2 has been used as the index of
differentiation. It is noticeable that the ferromagnesian major element oxides (FeO, MgO,
CaO) as well as TiO2 and P2O5 in the other three wells behave in a manner akin to well OW-
922 (Figure 23), showing a gradual decline with increasing SiO2 values. For K2O and Na2O,
the trend starts with a positive slope and bends to a negative one at ~ 65 wt.% and ~ 62 wt.%
SiO2 respectively. The change in slope is far more pronounced in Na2O than K2O. Similarly,
V, Sr, Sc and Co are anti-correlated with SiO2 (Figure 29), and show a very similar overall
picture to the ferromagnesian major oxides, pointing to their compatible behavior. On the
other hand La, Y, and Zr exhibit positive correlations with SiO2. It was observed that rocks
from wells OW-905A, 910 and 917 are slightly more enriched in SiO2 (44-80 wt.%) than
corresponding rocks from the study well (49-73 wt.%). In summary, comparison of the
geochemical results for the selected major and trace elements for the four wells demonstrate
a perfect overlap across the whole range of compositions.
0
4
8
12
16
40 50 60 70 80 90
FeO
(w
t.%
)
SiO2 (wt.%)
OW-922 OW-905A OW-910 OW-917
0
2
4
6
8
40 50 60 70 80 90
MgO
(w
t.%
)
SiO2 (wt.%)
OW-922 OW-905A OW-910 OW-917
0
4
8
12
16
40 50 60 70 80 90
CaO
(w
t.%
)
0
1
2
3
4
5
40 50 60 70 80 90
TiO
2(w
t.%
)
Figure 28: A comparison for variation in concentration of selected major element oxides
for wells OW-922, 905A, 910 and 917.
61
0
2
4
6
8
10
40 50 60 70 80 90
Na 2
O (
wt.
%)
SiO2 (wt.%)
OW-922 OW-905A OW-910 OW-917
0
2
4
6
8
40 50 60 70 80 90
K2O
(w
t.%
)
SiO2 (wt.%)
OW-922 OW-905A OW-910 OW-917
0
200
400
600
800
1000
40 50 60 70 80 90
Sr (
pp
m)
0
100
200
300
400
40 50 60 70 80 90
V (
pp
m)
0
10
20
30
40
50
40 50 60 70 80 90
Sc (
pp
m)
SiO2 (wt.%)
OW-922 OW-905A OW-910 OW-917
0
20
40
60
80
40 50 60 70 80 90
Co
(p
pm
)
SiO2 (wt.%)
OW-922 OW-905A OW-910 OW-917
Figure 29: A comparison for variation in concentration of selected trace elements for
wells OW-922, 905A, 910 and 917.
Figure 28 (continued): A comparison for variation in concentration of selected major
element oxides for wells OW-922, 905A, 910 and 917.
62
4.9 Microprobe analysis
Selected samples of epidote and chlorite from wells OW-922, 905A, 910 and 917 were
analysed with an electron microprobe (EMP). An important aspect of this analysis is to
assess the chemical composition and the compositional range of epidote and chlorite in the
mentioned wells. Interpretations made have been correlated with compositional ranges
observed in other geothermal systems (e.g. Bird et al., 1988; Marks et al., 2010; Lonker et
al., 1993; Bragason, 2012; Bird and Spieler, 2004). The first appearance of epidote and
chlorite has commonly been used as a criterion for setting production casing in production
wells in the Olkaria geothermal field. The selected epidote and chlorite samples analysed in
the present study occur either as open space fillings in vesicles and fractures or as a
replacement of original rock forming minerals, mostly plagioclase and sanidine. Associated
alteration minerals were separated and not simultaneously analysed alongside these two
minerals.
The mineral analyses were performed with a JEOL JXA-8230 electron microprobe located
at the Institute of Earth Sciences, University of Iceland using carbon coated finely polished
thin sections. The elements analysed were Si, Ti, Al, Fe, Mn, Mg, Ca, Na, K and Sr.
Quantitative analyses with wavelength-dispersive spectrometers (WDS) were performed at
an accelerating voltage of 15.0 keV, a cup current of 5.0 nA, and a beam diameter of between
5 to 10 µm. Counting time on peak was 40 seconds for all elements, except Na (30 seconds).
Background counting time varied for different elements. In the case of chlorite Si, Al, Fe,
Mn and Mg backgrounds were counted for 20 seconds, those of Ti, Ca, Sr and K for 40
seconds and Na 15 seconds. Background counting times for epidote analyses were 20
seconds for Si, Al, Fe, Mn and Ca, 40 seconds for Ti, Mg, K and Sr, and 15 seconds for Na.
Standards used were natural mineral standards. Structural formulae were calculated on the
basis of 24 oxygens. It should be noted that Fe is calculated as ferrous and ferric iron in
chlorite and epidote, respectively, and in all cases, Mn is calculated as divalent. Additional
data sets for this study are shown in Appendix D.
0
500
1000
1500
2000
2500
3000
3500
40 50 60 70 80 90
Zr (
pp
m)
SiO2 (wt.%)
OW-922 OW-905A OW-910 OW-917
0
100
200
300
400
500
40 50 60 70 80 90
Y (p
pm
)
SiO2 (wt.%)
OW-922 OW-905A OW-910 OW-917
Figure 29 (continued): A comparison for variation in concentration of selected trace
elements for wells OW-922, 905A, 910 and 917.
63
Epidote [Ca2FexAl3-xSi3O12(OH)] is an important index mineral in geothermal systems. The
first appearance of this mineral demonstrates convincingly temperatures equal to or in excess
of 240oC. Detailed analysis of epidote compositions has been carried out in a number of
geothermal systems, e.g. Salton Sea, Cerro Prieto, Los Azufres, Reykjanes, just to name a
few with high level of success. However, no quantitative analysis of epidote have been done
in the Olkaria geothermal system despite epidote being such an important mineral. The
ability to accurately determine element concentrations of epidote by EMP analysis is crucial
for a complete understanding of the chemical composition and changes in the compositional
range.
There are three octahedral sites, M(1), M(2) and M(3), in monoclinic epidote where
significant exchange of Fe3+ and Al3+ occurs. Ferric iron readily substitutes for Al3+ in both
the M(1) and M(3) sites. Several studies (e.g. Arnason et al., 1993) have indicated that Fe3+
is preferentially incorporated in the energetically larger and more distorted M(3) site and, to
a lesser degree, in the smaller and more regular M(1) site. At relatively low temperature, site
M(2) is nearly free of Fe3+ and mostly contains Al3+. It is only at high temperature that ferric
iron enters the M(2) site.
The compositional range of most natural epidotes in active geothermal systems is closely
described as a binary solid solution between the end-members clinozoisite [Ca2Al3Si3O12
(OH)] and pistachite [Ca2Fe3Si3O12(OH)]. Hence, the composition variation is dominated by
the substitution of Fe3+ and Al3+ in the octahedral sites (Marks et al., 2010). Epidote
composition is commonly expressed as the mole fraction of the pistachite component. The
notation Xps has been adopted to denote the mole fraction of pistachite [Xps = nFe / (nFe +
nAl)] where nFe and nAl designate cation proportions in each formula unit.
Epidote paragenesis is controlled by a variety of intensive and extensive thermodynamic
variables characterising hydrothermal systems. These include CO2 and O2 fugacities (ƒCO2,
ƒO2), coexisting mineral assemblages, changes in fluid chemistry, bulk rock composition
and temperature (e.g. Bird and Spieler, 2004). The functional relationships between epidote
solid solution compositions and changes in these variables depend on the stoichiometry of
reactions and enthalpies (Arnason et al., 1993).
Epidote was analysed in four different samples in this study; two samples derived from two
different depth intervals in well OW-910 and, one sample from a single depth range in each
of wells OW-905A and 922. No epidote sample was identified in well OW-917 for
microprobe analysis. It is important to note that, due to the paucity of epidote samples
analysed from wells OW-922 and 905A, it has not been possible to correlate the stratigraphy
in the three wells on the basis of major and minor element contents. Such comparison would
have been of great interest so as to enable critical assessment of the compositional range.
Therefore, the compositional range discussed in the succeeding sections is based largely on
analytical results from well OW-910. Epidote samples analysed were noted to occur in
association with chlorite, titanite, magnetite and albite (e.g. Figures 30 and 31). A
representative selection of chemical analyses illustrating the range in the composition of
epidote is shown in Table 6. As iron in epidote is dominantly in the trivalent state (Fe3+),
total Fe is reported as ferric iron (Fe2O3) as mentioned earlier. Low totals are at least partly
caused by the presence of the hydroxyl ion (OH−) in the epidote structure, which is not
detected by the microprobe.
64
Table 6: Representative EMP analyses of epidote from wells OW-922, 910 and 905A.
Well OW-922 OW-922 OW-910 OW-910 OW-910 OW-910 OW-905A OW-905A
Depth 2782 m 2782 m 864 m 864 m 1236 m 1236 m 1560 m 1560 m
Analysis as Oxides, wt.%
SiO2 37.48 37.50 37.32 38.14 36.16 35.77 37.12 36.39
TiO2 0.00 0.03 0.24 0.02 0.47 0.49 0.14 0.08
Al2O3 24.62 24.71 22.21 25.57 8.03 7.23 23.10 23.48
Fe2O3 12.71 12.48 16.14 11.86 19.91 21.11 14.86 13.78
MnO 0.11 0.17 0.14 0.04 0.48 0.39 1.53 1.05
MgO 0.11 0.05 0.09 0.06 0.35 0.29 0.05 0.02
CaO 22.97 22.81 23.21 23.15 32.59 33.05 21.50 22.16
Na2O 0.02 0.03 0.00 0.00 0.02 0.01 0.00 0.01
K2O 0.04 0.04 0.00 0.01 0.03 0.05 0.01 0.01
SrO 0.36 0.32 0.10 0.22 0.00 0.01 0.05 0.01
Total 98.43 98.13 99.46 99.08 98.04 98.41 98.35 97.00
Structural formulae based on 24 Oxygens
Si 5.694 5.707 5.681 5.721 5.961 5.916 5.693 5.647
Ti 0.000 0.003 0.028 0.002 0.058 0.061 0.016 0.010
Al 4.409 4.432 3.984 4.521 1.559 1.409 4.175 4.295
Fe 3+ 1.454 1.429 1.849 1.339 2.471 2.628 1.715 1.609
Mn 0.015 0.021 0.018 0.005 0.067 0.055 0.199 0.139
Mg 0.024 0.012 0.021 0.014 0.086 0.070 0.011 0.005
Ca 3.740 3.720 3.785 3.721 5.756 5.857 3.534 3.684
Na 0.006 0.010 0.000 0.001 0.005 0.004 0.000 0.002
K 0.008 0.007 0.001 0.002 0.006 0.010 0.001 0.002
Sr 0.032 0.028 0.009 0.020 0.000 0.001 0.004 0.001
Total 15.382 15.370 15.375 15.346 15.970 16.012 15.348 15.393
65
Figure 30: BSE image of epidote from 864 m in well OW-910. Compositional
heterogeneity is clearly seen in this figure with Al-rich sections appearing darker
and Fe-rich sections lighter in color. Numerous compositional zonation patterns
are represented by Al-rich rims and Fe-rich cores. Also, note the titanite inclusions
in epidote.
Figure 31: BSE image of epidote filling a vesicle from 1560 m in well OW-905A.
Epidote is observed occuring as aggregates of small columnar cystals. Quartz is
also noted growing in the vesicle. Black areas in the lower right hand corner, at
the centre and within quartz are epoxy.
66
The molar proportion of pistachite in epidote from the three wells ranges between 0.23-0.65,
as shown in the table below.
Table 7: Range and average molar proportion of pistachite (Xps) component in epidote.
Well Depth (m) Molar ratio of pistachite
nFe / (nFe + nAl)
range average
OW-922 2782 0.23 – 0.24 0.235
OW-910 864 0.23 – 0.32 0.275
OW-910 1236 0.61 – 0.65 0.630
OW-905A 1560 0.29 – 0.27 0.280
Chemical analyses presented in Table 6 indicate that the epidotes are relatively Al-rich types,
with the exception of epidote from 1236 m in well OW-910, which is Fe-rich. There is a
slight inhomogeneity in composition with respect to the Fe3+/Al3+ ratio of the epidotes
analysed. A narrow intragrain chemical compositional change exists for the major and minor
oxides for the three wells. However, intergrain chemical compositional change is more
pronounced in well OW-910, based on two samples derived from two different depth
intervals (864 m and 1236 m). Here, a systematic change in the composition of epidote with
increasing depth is clearly noticeable. This is illustrated by the change in the contents of the
two dominant elements, i.e. Al and Fe, that predominantly dictate the variation in the
composition of epidote (Figures 32 and 33). Al2O3 contents show a significant decline from
~ 24 to ~ 8 wt.% on average, whereas, Fe2O3 contents display a gradual increase from 14 to
~ 21 wt.% on average with depth. This is consistent with increasing average molar ratio of
pistachite component with depth (Table 7). Compositional zoning, caused by variable Fe3+-
Al3+ substitution and representing phase disequilibrium, is a common feature of epidotes
analysed in geothermal systems (Arnason and Bird, 1992). Other minor elements of
importance observed to be in variable concentration in epidote, though, to a lesser extent,
are Sr and Ti. These elements, like Al and Fe, are enriched on the rims and cores of epidote,
respectively, and are clearly visible on intragrain zonation.
67
Figure 32: A WDS elemental map showing typical compositional zoning of epidote from
864 m in well OW-910. Grayscale map (a) represent the original composition, whereas,
maps (b), (c) and (d) denote Fe, Sr and Al levels respectively.
68
Chlorites are basically hydrous silicates of aluminium, magnesium and/or iron, represented
by the structural formula (Mg,Al,Fe)12 (Si,Al)8 O20 (OH)16. Common cation substitution
involves replacement of Si4+ by Al3+ in a tetrahedral site, and, a limited substitution of R2+
by Al3+ in the octahedral site to preserve charge balance, where R2+ essentially is a
summation of Mg2+ + Fe2+ + Mn2+. Analyses of chlorite for this study were made from six
samples, two each derived from different depth intervals in wells OW-922 and 910. From
wells OW-905A and 917, one sample was analysed, representing a single depth range in
each case. Representative chemical analyses of chlorite are presented in Table 8. A review
of chlorite analyses (e.g. Dubois and Martinez-Serrano, 1998) shows that ferric iron typically
constitutes < 5% of total Fe. Therefore, it is appropriate to report Fe as divalent.
It should be noted that chlorite has a strong tendency to form very fine-grained mats of
needle-shaped mineral grains (e.g. Figure 34). This along with the relatively rough, uneven
surface of the chlorite mats results in relatively poor analyses. As chlorite is highly hydrated,
totals will always be low, but this can also mean that the mineral is relatively unstable under
the electron beam, even with a relatively low probe current and wide beam diameter. Hence,
the results of the EMP analyses of chlorite have to be interpreted with some caution. Most
Figure 33: A WDS elemental map showing Fe (b) and Al (d) distribution within a single
epidote grain from 2782 m in well OW-922. The darker and lighter sections in the original
compositional map (a) indicates Al and Fe-rich regions, respectively. Map (c) shows Sr
levels. Compositional zoning is not obvious as is the case in Figure 32.
69
published studies (e.g. Lonker and others, 1993) have shown that variations in the chemical
composition of chlorites are largely dependent on temperature and changes in the
composition of the host rocks. Changes in ƒO2 of geothermal fluids at depth is another
important factor which cannot be discounted, even though it plays a limited role. Variations
in host rock composition mainly affects distribution of Fe in chlorite phases, consequently
producing some changes in Fe-Mg occupancy (Bryndzia and Scott, 1987).
Table 8: Representative EMP analyses of chlorite for wells OW-922, 910, 917 and 905A.
Well OW-922 OW-922 OW-910 OW-910 OW-917 OW-905A
Depth 2236 m 2782 m 864 m 1236 m 1870 m 1560 m
Analysis as Oxides, wt.%
SiO2 22.20 22.12 25.02 30.45 19.58 25.61
TiO2 0.10 0.06 0.03 0.11 0.04 0.00
Al2O3 14.86 13.85 12.95 16.90 16.33 19.16
FeO 30.50 32.97 12.60 17.62 34.55 34.41
MnO 0.12 0.33 0.19 0.28 0.70 2.02
MgO 5.38 5.38 17.58 15.59 5.30 7.87
CaO 0.10 0.16 0.31 0.44 0.07 0.11
Na2O 0.75 0.18 0.03 0.09 0.47 0.02
K2O 0.32 0.13 0.02 0.01 0.58 0.02
SrO 0.03 0.00 0.00 0.02 0.00 0.02
Total 74.35 75.18 68.73 81.50 77.62 89.26
Structural formulae based on 24 Oxygens
Si 5.050 5.150 5.446 5.634 4.417 4.821
Ti 0.017 0.011 0.006 0.015 0.007 0.000
Al 3.985 3.727 3.323 3.686 4.342 4.250
Fe2+ 5.803 6.296 2.294 2.726 6.518 5.418
Mn 0.023 0.064 0.035 0.043 0.133 0.322
Mg 1.824 1.831 5.703 4.298 1.783 2.209
Ca 0.024 0.039 0.073 0.087 0.017 0.023
Na 0.329 0.080 0.013 0.031 0.207 0.009
K 0.092 0.039 0.005 0.004 0.167 0.005
Sr 0.004 0.000 0.000 0.002 0.000 0.003
Total 17.151 17.236 16.898 16.526 17.591 17.060
70
Chemical analyses of chlorites from the four wells indicate that they are Fe-rich varieties
except from 864 m depth in well OW-910 (Table 8). Most of the chlorites fall in the
brunsvigite compositional field going by the classification scheme proposed by Dubois and
Martinez-Serrano (1998). Si, Al, Fe and Mg are in appreciable amounts in chlorite in the
four wells, whereas Ti, Mn, Ca, Na, K and Sr are present as minor or trace elements.
Variations in Fe2+ / (Fe2+ + Mg) ratio in chlorite from hydrothermally altered rocks (Table
9) has in many cases (e.g. Lonker and others, 1993) been applied to infer permeable zones
in a well.
Table 9: Molar ratio of chlorite from wells OW-922 and 910.
Well Depth (m) Molar ratio of chlorite
nFe / (nFe + nMg)
OW-922 2236 0.76
OW-922 2782 0.77
OW-910 864 0.29
OW-910 1236 0.39
Figure 34: BSE image of chlorite replacing plagioclase at 1870
m in well OW-917.
71
5 Discussion
5.1 Lithological units
Well OW-922 is a step-out vertical well drilled to a total depth of 2990 m CT to the east, in
the plain between Olkaria Domes field and Longonot volcanic centre. The main purpose of
drilling the well was to appraise the geothermal resource potential in the area, so as to expand
production drilling outside the already congested proposed Olkaria caldera structure. This,
also comes at a time of renewed focus by the Kenyan government on accelerating
development and utilisation of geothermal resources for electricity generation in the country.
Based on analysis of drilling cuttings through binocular and petrographic microscopy and
ICP-OES, the sub-surface strata penetrated by the investigation well comprise pyroclastics,
tuffs, rhyolite, trachyandesite, basaltic trachyandesite, basalt, trachydacite and trachyte.
From the existing geologic well logs (KenGen in-house data), it is evident that no other well
drilled in Olkaria has intersected such thick sequence of pyroclastic deposits, extending
continuously for about 300 m, as is the case with the study well. The observation is perhaps
caused by the well’s proximity to the Longonot volcanic complex.
Stratigraphically, the Longonot pyroclastics (light grey to grey in colour) overlie the Olkaria
pyroclastics (yellowish brown in colour), attesting that the former are younger in age. Like
in most Olkaria wells, trachyte is still the dominant rock unit in the study well, in some cases
extending steadily downward to reach a thickness of about 1200 m (e.g. between 1700-2900
m). The unit forms the reservoir rock in the study well. Clearly, the rhyolites are markedly
less abundant in the study well relative to wells OW-905A, 910 and 917, perhaps caused by
the well's location being outside the main Olkaria volcanic zone. As a result, rhyolitic
volcanism seems to have been majorly confined to the proposed Olkaria caldera structure.
No intrusive rocks were intersected by the well. Intrusives result from the injection of magma
into the overlying country rock and commonly serve as a source of heat to shallow levels of
the geothermal system. Based on this observation, it is probable that pre- and post-caldera
magmatism associated with the OVC had minimal effect in the area surrounding the well.
Lack of intrusive rocks in the well further attests to the location of the well being outside the
main Olkaria volcanic zone.
5.2 Hydrothermal alteration
Hydrothermal alteration sequences provide a significant insight into the history of a
geothermal system. Hydrothermal alteration observed in the study well is essentially made
up of two components: (a) replacement of primary components in the rocks by alteration
minerals, and (b) precipitation of alteration minerals into open spaces (veins, fractures and
vesicles) in the rock. Generally, there is scarcity of high-temperature alteration minerals in
well OW-922. The top most zone of the well (0-190 m) contains relatively fresh rocks,
including a sequence that do not show any indication of geothermal interaction and at the
most show moderate oxidation due to groundwater circulation. The subsequent zone
between 190 and about 450 m is dominated by relatively low-temperature alteration mineral
assemblages (e.g. zeolites, smectites, chalcedony), signifying that inferred alteration
temperatures are < 200oC. Petrographic examination of the rock samples between this
interval indicate that the primary phases replaced are mostly volcanic glass and olivine, as
seen in Table 3. A prograde mineralogical change, marked by transition from secondary
quartz, wairakite, albite, illite, chlorite and epidote is evident between 476 m to about 1798
72
m depth. The occurence of epidote, which is regarded as one of the most important
greenschist alteration mineral assemblages (e.g. Bird and Spieler, 2004) is relatively
spasmodic. In most cases, the occurence of the mineral is observed as faint colouration,
possibly reflecting that equilibrium between the host rocks and reservoir fluids was seldom
achieved. It is worth noting that certain minerals, including calcite and pyrite form at both
low and high temperatures (i.e. between ~ 50-300oC), thus cannot be totally relied upon as
accurate temperature indicators (Brown, 1978). Below 1038 m depth to the well bottom,
mixed layer clays (MLCs), occuring in different combinations (smectite/illite,
smectite/chlorite and smectite/illite/chlorite), and calcite are the dominant alteration mineral
phases.
Clay mineralogy in the study well differs somewhat from clay alteration commonly observed
in other Olkaria wells. In particular, the persistence of MLCs to the bottom of the well
possibly reflects modification of the composition of hydrothermal fluids. This is in
agreement with reports of various authors (e.g. Steiner, 1977; Kristmannsdóttir, 1979) who
found out that the distribution of clay minerals in geothermal areas depends on the fluid
chemistry and thermal structure of the field. Studies of chemical composition of the fluids
are, however, required to prove the case. In previous studies in Olkaria (e.g. Mwania et al.,
2013; Musonye, 2015), it is observed that the occurence of MLCs is mostly restricted to
depths above 1500 m. MLCs in this well are characterized by well developed smectite peaks
at greater depths, for example at 2616 m, 2840 m and 2918 m depth, just to mention a few,
perhaps indicating that temperatures are not high enough. The association of MLCs with
haematite and abundant calcite, particularly towards the bottom of the well could reflect that
temperatures are relatively low in this region.
5.3 Permeability and comparison of
temperatures
Feed zones below the production casing in a well can either be cold or hot. Cold feed zones,
depending on the size, may ultimately cool the well, rendering it less productive. On the
other hand, hot feed zones in many cases are known to maximise well productivity. In the
study well, feed zones inferred demonstrate a relationship between geological observations,
variations in drilling parameters and kinks in temperature recovery logs. However, it is key
to point out that intrusion related feed zones, which are common in wells OW-905A, 910
and 917 are not present in well OW-922 as already discussed. Based on analysis of the feed
zones in relation to the indicators mentioned above, it is noted that most of the feed zones
are small, leading to the conclusion that permeability in the well is limited. This observation
is also supported, for example by the failed injection test where the well did not accept water,
absence of clear pivot point and total failure of the well to discharge after 49 days of heat-
up. The pivot point is fundamental in the sense that it is ordinarily used to infer the best feed
zone in a well. At this point, the pressure is nearly constant. Moreover, it is probable that
intense calcite deposition, particularly towards the bottom of the well, clogged the once
present primary and secondary fluid channels. It is also observed that a correlation between
secondary mineralisation and identified feed points is restricted, and the values of key
indicators of water/rock interaction (i.e. Na2O and K2O) do not show any large shift within
and in the vicinity of the feed zones as would be expected.
73
A comparison of measured formation and alteration temperatures (Figure 22) indicates that
the geothermal system around the study well has substantially been disturbed. It is apparent
that the inferred alteration temperature plots higher than the measured formation
temperature. This demonstrates that past temperatures were substantially higher than the
present measured ones. Temperature profile argues in favour of this observation, showing a
maximum temperature of 127oC betweeen ~2400- 2500 m depth (Figure 21), below which
there is a reversal in temperature gradient to the well bottom. This is inconsistent with the
inferred alteration temperature of about 240oC, denoted by the presence of epidote. It is
necessary though to mention that the measured formation temperature was taken only after
49 days recovery period. As a result, the formation temperature may not be representative
enough, since the geothermal system had likely not fully stabilized following the long
drilling time of the well. An up to date temperature log would provide a true reflection of
the current state of formation temperature in the well. It is, therefore, likely that significant
cooling has occurred in the geothermal system around well OW-922 over time.
Other possible indications of cooling in the well is demonstrated by the extensive deposition
of calcite (e.g. Figure 21), especially towards the well bottom (i.e. ~ 2240-2990 m). A
reasonable explanation for the intense calcite deposition, which is consistent with the
argument of Simmons and Christenson (1994), implies heating of the colder peripheral fluids
entering the geothermal system by the host rocks, consequently favouring carbonate
precipitation. This observation is also in accord with the interpretation put forward by
Musonye (2015) to explain intense calcite deposition in well OW-917. The latter was drilled
along the proposed caldera structure (Figure 3) and exhibits more or less similar
hydrothermal alteration features to well OW-922. Cases of calcite superimposed on epidote
(e.g. at 2782 m depth) has been observed by several authors (e.g. Mortensen et al., 2014;
Franzson et al., 2008) and interpreted to reflect carbonate deposition during cooling of the
system by influx of colder fluids into the reservoir. However, fluid inclusion studies would
be necessary to provide detailed insights about the temporal evolution of the geothermal
system around the study well.
Equally, haematite is observed in moderate amounts (i.e. at level 2 on the scale of 1-3)
between 2342 and 2982 m depth. In this case, the occurence of haematite can be interpreted
to suggest influx of oxygen-rich cooler groundwater into the system. The geothermal system
around the study well can therefore, be described as episodic rather than continuous. An
episodic geothermal system refers to a system where some changes (either heating or
cooling) has occured over time based on differences between inferred alteration and
measured (or calculated) formation temperatures. On the other hand, a continuous system is
where the inferred alteration and measured (or calculated) formation temperatures are more
or less in a state of equilibrium (Brown, 1978).
74
Figure 35: A possible conceptual model based on outcome of the study, along the
traverse A-B.
75
5.4 Rock composition
Rocks from the study well show a complete range of compositions from basalt to trachyte
or rhyolite. The dominance of evolved rock suites relative to mafic and intermediate rocks
implies high degree of differentiation of the parental magma. Peccerilo et al. (2007) observed
that the relative scarcity of the intermediate rocks are related to their high viscosity, which
inhibits them from erupting. Even though the more silicic magmas (pantellerites and
comendites) still have high viscosities, they tend to be associated with much lower density.
This feature, alongside the high volatile contents (e.g. CO2, H2O, Cl and F) of silicic
magmas, helps to drive them out after high degrees of fractional crystallisation from the
magma reservoirs (Ronga et al., 2009). Based on the TAS classification scheme, the
prevalence of trachytes is in accord with the general stratigraphy of Olkaria, which is
composed primarily of trachytes, except for the Olkaria west fields.
5.5 Hydrothermal alteration effects on rock
chemistry
Interaction of high-temperature fluids with host rocks (hydrothermal alteration) in the
geothermal system around well OW-922 has produced changes in the rocks. Essentially,
hydrothermal alteration is a very complex process involving a chemical, mineralogical and
textural changes of the host rock. The process results from the interaction of the wall rock
with hot aqueous fluids (Pirajno, 1992). These fluids deliver the chemical reactants and take
away the aqueous reaction products. The complex water-rock interaction process has
undoubtedly changed contents of the chemical elements in the altered rocks. In this case, a
direct method, based on comparison of elemental concentrations has been applied to
determine the effects of hydrothermal alteration on selected major and trace elements. Even
though this method serves the purpose to some extent, a distinct disadvantage is that it does
not put reliable constraints on the effect of alteration on element mobility in altered rocks.
Other methods, such as the statistical approach (e.g. Student t-tests) and mass balance
calculations, are known to yield more coherent results. Quantification of alteration effects
on an element is based on the concept of element mobility. The amount of hydrothermal
alteration is very often correlated with an element’s ionic potential (charge/radius ratio) (e.g.
Verma et al., 2005). Consequently, highly mobile elements are characterised by low ionic
potential (< 0.03 pm-1) and tend to be easily leached away by the geothermal fluids as
hydrated cations. These elements exhibit large variations (scatter) as a consequence of the
alteration process and are exemplified here by Na, K, Ba, Sr, and Rb. In contrast, most
immobile elements are characterised by high ionic potential (> 0.10 pm-1) and tend to be
retained in the host rocks since they are not easily leached away by hydrothermal fluids. In
this study, these elements include Ti, P, Zr, Y, Sc, La and the transition metals (Co, Cr, Ni,
V).
5.5.1 Effects on major element geochemistry
Selected major elements have been plotted against SiO2 to decipher alteration effects on
major element geochemistry (Figure 36). SiO2 has been used as the reference or independent
variable because it is a nearly conservative element (oxide) during hydrothermal alteration
processes. This view has been advocated by Verma et al. (2005), who have presented
76
convincing experimental and theoretical results. The authors suggest that SiO2 contents of
most geothermal waters are generally low (< 0.15%). As a result, the SiO2 contents of the
altered rocks should be more or less unaffected, and hence, the use of Si or SiO2 as a
conservative independent element or oxide for assessing alteration behavior of other
elements (or element ratios) is widely accepted.
The variation of selected major elements vs. SiO2 indicates that hydrothermal alteration has
had insignificant effect on Al2O3 and MnO in comparison with Na2O and K2O. Na2O and
K2O display some scattering, which is more pronounced in K2O and, are traditionally known
to be highly mobile during hydrothermal alteration (e.g. Verma et al., 2005). These elements
are the dominant constituents in alkali feldspar, which is a primary rock forming mineral in
trachytes and rhyolites. Consequently, this raises the possibility that, as the two elements are
highly mobile, they have been affected by post-magmatic alteration processes. The
preferential mobility of the alkalis compared to aluminium during hydrothermal alteration,
could also be responsible for the marked metaluminous character of a majority of the
samples in the study well. This is expressed by molar proportions Al2O3 > Na2O + K2O. The
cause of scattering of MnO is not intuitively obvious, but seems to be related to variable
degrees of fractionation as opposed to a true scatter. Another possible explanation could be
the result of background noise introduced by using analyses of highly porphyritic samples
(e.g. Wilson, 1989).
1
2
3
4
5
6
7
8
45 50 55 60 65 70 75
Na 2
O (
wt.
%)
6
8
10
12
14
16
18
Al 2
O3
(wt.
%)
0
2
4
6
8
45 50 55 60 65 70 75
K2O
(w
t.%
)
SiO2 (wt.%)
0,0
0,1
0,2
0,3
45 50 55 60 65 70 75
Mn
O (
wt.
%)
SiO2 (wt.%)
Figure 36: Variation diagrams showing the concentrations of Na2O, Al2O3, K2O and
MnO against SiO2.
77
5.5.2 Effects on trace element geochemistry
A plot of selected trace elements against Zr (Figure 37) have been used to assess the
alteration effects on trace element geochemistry. Zr has been used as an independent variable
owing to its large range of variation and being regarded as a highly immobile element in
volcanic rocks (Franszon et al., 2008).
From the selected trace element vs. Zr. plots, it is evident that hydrothermal alteration has
the least effect on Sr, Ni and Ba, but has severely mobilised Rb, as shown by large scattering.
La and Y are comparatively inert and not much affected by hydrothermal alteration
processes. In previous studies (e.g. Verma et al., 2005), the two elements have been
considered as typical fluid-immobile elements. Rb is an alkali metal and, is a major
constituent of alkali feldspars. The element is highly mobile and is easily leached from the
0
100
200
300
400
0 500 1000 1500 2000 2500 3000
Sr (
pp
m)
0
50
100
150
200
250
0 500 1000 1500 2000 2500 3000
Rb
(p
pm
)
0
100
200
300
400
500
0 500 1000 1500 2000 2500 3000
La (
pp
m)
0
20
40
60
80
100
120
0 500 1000 1500 2000 2500 3000
Ni (
pp
m)
0
500
1000
1500
2000
2500
0 500 1000 1500 2000 2500 3000
Ba
(pp
m)
Zr. (ppm)
0
100
200
300
400
0 500 1000 1500 2000 2500 3000
Y (p
pm
)
Zr. (ppm)
Figure 37: Variation diagrams showing the concentrations of Rb, Sr, La, Ni, Ba and Y
against Zr.
78
rocks by fluid components of the hydrothermal alteration process. Sr, even though normally
considered mobile, has been least affected by alteration.
It must be emphasised that the ICP-OES analysis did not include measurement of loss on
ignition (LOI). The latter has been used in many cases (e.g. Franszon et al., 2008) to monitor
alteration. Highly altered rocks tend to have high LOI values, whereas, the unaltered or
slightly altered rocks are characterised by low LOI values. In addition, Verma et al. (2008)
observed that high LOI are associated with low Na2O abundances, strongly indicative of Na
loss during alteration processes.
5.6 Magmatic differentiation processes
Magma differentiation or segregation processes are central to understanding the origin of
sequences of volcanic rocks exhibiting systematic variations in composition, despite being
obtained from a likely analogous source component. Insights into these processes are derived
from studies of lavas or intrusions exposed on the surface or obtained through drilling of the
sub-surface geological formations, as is the case in the present study. Importantly, the rocks
provide excellent opportunity of tracing changes in major and trace elements during
magmatic evolution. Models of magma differentiation, including magma mixing/mingling,
crustal contamination, partial melting and fractional crystallisation, have been tested by
many authors (e.g. Wilson, 1989; Omenda, 1997; Peccerillo et al., 2007).
To decipher the differentiation mechanisms for the sub-surface rocks in well OW-922, SiO2
was chosen as the abscissa or index of differentiation owing to its abundance and wide
concentration range (59-73 wt.%). Consequently, it can be used to observe the progress of
differentiation processes. In spite of the wide range of SiO2 contents, it is pertinent to note
that majority of the samples have silica values clustering between 63 and 73 wt.%. All the
major oxides and trace elements have been plotted against SiO2 on Harker diagrams (Figures
23 and 24) and the interpretations discussed below.
5.6.1 Major elements
Progressive depletion in major element oxides, such as FeO (total iron expressed as ferrous
oxide), MgO, CaO, TiO2 and P2O5 show curved to near-curved trends in the variation
diagrams. Whole-rock SiO2 contents are negatively correlated with these elements,
indicating that simultaneous fractional crystallisation process of mafic mineral assemblages
from a basaltic parental magma is a possible cause. The inverse trend observed between FeO
and MgO vs. SiO2 (Figure 23) can be interpreted as implying the separation of
ferromagnesian mineral phases (olivine and clinopyroxene) from the melt during early stages
of crystallisation (e.g. Harðarson, 1993). Similarly, the decrease in CaO content strongly
suggests that the mafic to intermediate magmas evolved by fractionation of an assemblage
dominated by plagioclase and clinopyroxene.
The interpretation of TiO2 vs. SiO2 could be related to fractionation of Fe-Ti oxides (e.g.
ilmenite and/or titano-magnetite) from the intermediate to silicic rock suites. According to
Wilson (1989), TiO2 is incompatible in the main minerals of basalt. As a result, its
concentration generally increases significantly in the residual liquid after crystallisation of
olivine and plagioclase during early stages of crystallisation. The decrease in P2O5 as SiO2
content increases is ascribed to the beginning of apatite crystallisation in the intermediate to
silicic rocks, resulting in a less phosphor content of the melt.
79
The down-turn in Na2O abundances in samples containing greater than 64 wt.% SiO2 (Figure
23) denotes the point at which plagioclase fractionation becomes a controlling factor in the
compositional evolution of the residual liquid. This observation is consistent with the gradual
decline in Al2O3 contents, which would only require fractionation of plagioclase, and to a
lesser extent alkali feldspar (Weaver et al., 1972). The CaO/ALK (ALK = total alkali) ratio
for the analysed samples decreases markedly as SiO2 content increases (Figure 22). This
trend is consistent with a fractional crystallisation model (Weaver et al., 1972). In general,
the different slope of trends for the major oxides vs. SiO2 is likely influenced by the
fractionating minerals and their proportions. Low P.I. (0.26-1.03) of the samples analysed
could probably be attributed to significant loss of the alkalis which are easily leached away
during hydrothermal alteration processes as discussed earlier. Based on this observation, it
is reasonable to suppose that the loss of Na2O and K2O in this well was sufficient enough to
make all but two samples (i.e. 1106 and 1142 m) appear metaluminous (i.e. Al2O3 > Na2O
+ K2O).
5.6.2 Trace elements
The relative enrichment and/or depletion of trace elements offers valuable insights about the
source characteristics or petrogenetic process of any volcanic rocks suite. It is important to
note that trace elements do not form any major rock forming minerals themselves but mostly
enter crystallographic sites in the most common mineral phases by substitution. Various
mineral phases may either accept (compatible) or largely exclude (incompatible in such
phases) these elements into their crystal structure. Consequently, the use of trace elements
to diagnose petrogenetic processes leads to better understanding of the nature and
composition of the mineral assemblages with which the magma may have previously
equilibrated. Although the trace elements graphics generally show more scattering (Figure
24) compared to the major element oxides (Figure 23), systematic variations with SiO2
contents are clearly evident.
The decrease in Co, Cr, Ni and V with increasing SiO2 content (Figure 24) can be
interpreted to signify fractional crystallisation of mafic mineral assemblages (olivine and
clinopyroxene) from a basaltic parental magma. From the description of the lithologies in
well OW-922, it is noticeable that olivine and clinopyroxene are frequently observed as the
dominant phenocryst phases in majority of the mafic rocks. Nickel partitions easily into
olivine while V is mainly incorporated into clinopyroxene in mafic rocks (Wilson, 1989).
The depletion of Sc towards more evolved rocks is consistent with the fractionation of
olivine and plagioclase mineral phases. Cobalt follows similar pattern as Sc and is known
to partition strongly into olivine (Weaver et al., 1972). Extraction of Cu is usually not
related to any precipitating major mineral phases but can be linked to the separation of a
sulphide minerals (Weaver et al., 1972).
The main phase affecting Sr partitioning in the mafic rocks is plagioclase and to a lesser
extent alkali feldspar. Therefore, the systematic enrichment in Sr contents in the mafic and
intermediate rocks can be accounted for by the accumulation of plagioclase (and possibly
alkali feldspar). Additionally, incorporation of Sr into apatite or titanite may as well
contribute to its reduction in the more evolved lavas (Weaver et al., 1972). However, most
of the trend observed in Sr. vs. SiO2 plot can be attributed to variable degrees of plagioclase
fractionation. ITEs, e.g. Y, Zr, Rb and La, show a relatively good positive correlation with
increasing SiO2 contents despite some scattering (Figure 24). The correlation in Rb vs. SiO2
can be termed strong correlation, whereas for Y, Zr and La vs. SiO2 is a moderately strong
correlation. Increase in ITEs concentrations with increasing SiO2 content can satisfactorily
80
be interpreted to mark their incompatible behavior, where the melt become steadily
enriched in these elements as fractionation proceeds.
The strong depletion of Ba (< 20 ppm) in rhyolites can be attributed to fractionation of
sanidine. The element becomes compatible when alkali feldspar appears in the phase
assemblage of intermediate and silicic melts. Here, it easily substitutes for potassium in K-
feldspar, specifically the sanidine. The latter is one of the prominent phenocryst phases in
the rhyolites. It is important to mention that fractionation of sanidines also ultimately
affects the concentration of potassium in the residual liquid. On the other hand,
accumulation of sanidines could possibly account for Ba enrichment in the trachytes. The
differentiation trends seen in the major and trace element systematics (Figures 23 and 24)
are consistent with a fractional crystallisation process for the evolution of the rocks in well
OW-922.
5.6.3 Assessing other differentiation mechanisms
To assess other petrogenetic processes responsible for the generation of these rocks besides
fractional crystallisation, plots of incompatible vs. incompatible elements are shown below.
Zr is used as an independent variable owing to its large range of variation, high
incompatibility and relatively immobile nature during hydrothermal alteration (Franzson
et al., 2008).
As discussed earlier, fractional crystallisation is the most viable model for the generation
of the evolved rocks in this well. This has been emphasized further by the La and Y vs. Zr
plots. A basic requirement for fractional crystallisation processes for any two ITEs is that
the two elements must plot on linear trends which can be extrapolated to the origin
(Omenda, 1997). In his pronouncement, fractional crystallisation does not significantly
change ratios of trace elements with analogues degrees of incompatibility. This
requirement has been successfully fulfilled by La and Y vs. Zr plots. However, fractional
crystallisation cannot be single-handedly put forth to explain all the geochemical features
depicted by La and Y vs. Zr plots.
0
100
200
300
400
500
0 500 1000 1500 2000 2500 3000
La (
pp
m)
Zr. (ppm)
0
100
200
300
400
0 500 1000 1500 2000 2500 3000
Y (p
pm
)
Zr. (ppm)
Figure 38: A plot of La vs. Zr. and Y vs. Zr. showing different magmatic differentiation
processes. The dotted and continuous lines indicates fractional crystallisation and other
petrogenetic processes, respectively.
81
Deviations from the continuous trends in the incompatible element plots (La/Zr, Y/Zr)
observed in the rhyolites suggest that other processes (e.g. through partial melting of crustal
rocks or magma mixing) at least contribute to their petrogenesis. These deviations are
exhibited predominantly by the rhyolites between 1106-1142 m and 2420-2570 m depth.
Previous studies of the magmatic differentiation mechanisms for the surface rhyolites in
Olkaria (e.g. Macdonald et al., 2008) agree with the observation made in the present study.
In their study, the authors established that rhyolites were formed primarily by fractional
crystallisation of metaluminous trachytic magmas and to a lesser extent by partial melting
of syenites. The trachytes themselves, were formed by the fractional crystallisation of
basaltic magmas through mugearites and benmoreites (Marshall et al., 2009).
It is therefore possible that rhyolites plotting off the main trend were generated by partial
melting of the syenites. However, no clear theory based on geochemical grounds has been
put forward by the authors to exhaustively explain the origin of rhyolites by partial melting,
even though there is a high probability that the process was promoted by fluxing of volatiles
emanating from underlying magmas. Evaluating the degree of partial melting is, however,
not feasible, because doing so requires knowledge of partition coefficients for the relevant
mineral phases, source compositions and the depth at which melting occurs. These are
rather poorly constrained.
5.7 Comparison of surface and sub-surface
samples
A comparison of surface samples from Olkaria and neighboring minor eruption centres and
sub-surface samples from the study well, and wells OW-905A, 910 and 917 (Figure 27)
clearly points towards a common petrogenetic mechanism for these rocks. This is exhibited
by both samples plotting within similar fields on the TAS diagram (Figure 27). On this
account, it is reasonable to suggest that magma from these volcanic centres was generated
by a common differentiation process, possibly from a similar parental magma. Only detailed
radiogenic isotope studies can convincingly prove the case. Correspondingly, comparison of
the geochemical data for selected major and trace elements for the four wells (Figures 28
and 29) demonstrate a perfect overlap across the whole range of compositions. A more
realistic interpretation would be that the trends displayed by the two groups of elements are
related by a common evolution mechanism.
These results are in agreement with the interpretations of Musonye (2015), who presented
convincing results in his study of the petrochemistry of wells OW-905A, 910 and 917. The
author construed high degree of correlation between certain major and trace elements vs. an
evolution index as being consistent with predominantly extended fractional crystallisation,
or even requiring fractional crystallisation as a necessary condition. Since the rocks from
well OW-922 are geochemically comparable with rocks from wells OW-910, 917 and 905A,
a petrogenetic connection seems likely between the rocks in the four wells.
Whilst these relationships are consistent with a fractional crystallisation process, it is not
necessarily the only mechanism that accounts for all the geochemical characteristics
observed. Other differentiation processes cannot be completely ruled out, however, limited
role they play in the evolution of these rocks. To illustrate this point, plots of two ITEs
plotted against each other have been made, as in Figure 38. Results of these plots (Figure
39) coincide with interpretations made for Figure 38. Once again, the unanswered question
82
remains whether these rocks are derived from similar or different parental magma. Once
again, only detailed radiogenic isotope studies can resolve the impasse.
5.8 Compositional variation of alteration
minerals
5.8.1 Epidote
Measured chemical compositions of different epidote crystals, derived from two different
depth intervals in well OW-910, show wide and concomitant variations in Al and Fe
contents. Variations in the Fe3+-Al3+ contents of epidote observed presumably reflect
changes in fluid composition during the formation of the mineral. Epidote composition
depends directly on the ratio of the activities of Fe3+and Al3+ in the fluid. However, aqueous
complexation of Fe3+ and Al3+ with other anions (e.g. SO42−, CO3
2−, Cl−, OH−, etc) may cause
the composition of epidote to depend on ƒCO2, pH and chloride concentration in the fluid
phase (Arnason et al., 1993). Thus, changes in fluid composition during the evolution of
epidote may result in variations in Fe3+and Al3+ exchange without necessarily reflecting
changes in temperature or pressure.
Other plausible thermodynamic mechanisms to account for the variation in epidote
composition in well OW-910 are changes in ƒO2 and the oxidation state. Diverse
experimental results (e.g. Arnason et al., 1993) have demonstrated that ƒO2 strongly controls
the Fe-content of epidote. Values for ƒO2 for the Olkaria geothermal system have not been
determined, but such studies would be of great interest. According to the authors, epidote is
the most Fe-rich in the presence of haematite-magnetite oxygen buffer and becomes more
aluminous with decreasing ƒO2, as the proportion of ferric iron in the system decreases. In
addition, the authors noted that epidote stability increases with increasing ƒO2. The latter
expands the P-T boundaries of the epidote stability field. Like in the Salton Sea geothermal
system, epidote in well OW-910 is known from binocular observation and optical
petrography of cuttings to occur alongside haematite. This observation is closely compatible
0
100
200
300
400
500
0 1000 2000 3000
La (
pp
m)
Zr (ppm)
OW-922 OW-905A OW-910 OW-917
0
100
200
300
400
500
0 500 1000 1500 2000 2500 3000
Y (p
pm
)
Zr (ppm)
OW-922 OW-905A OW-910 OW-917
Figure 39: Harker's variation diagrams between various pairs of incompatible trace
elements displaying fractional crystallisation (dotted line). The flexure from the main
trend (continuous line) represent other magma differentiation processes.
83
with the relatively high ƒO2 calculated for the Salton sea geothermal system (Bird and
Spieler, 2004). Epidote in haematite-bearing assemblages is traditionally known to be more
Fe-rich (Bird et al., 1988).
Additionally, results of epidote analysis at 1236 m depth display a positive relationship
between the relatively high Fe-content of epidote and inferred level of oxidation. This is
marked by concomitant increase in the oxidation state of host rocks from 864 m to 1236 m
depth. The enrichment of Sr and Ti in the rim and core of epidote, respectively suggest that
epidote could be an important repository for these elements, mobilized through hydrothermal
alteration. Results of previous analyses of Icelandic epidotes, for example in the Hengill
geothermal system (e.g. Bragason, 2012), points to a more or less similar picture where, Sr
contents are noted to increase progressively towards the rims. Based on the authors'
interpretation, increase in Sr contents very likely reflects extensive breakdown of
plagioclase.
Variations of the average mole fraction of pistachite (Xps) contents of epidote in this well is
satisfactorily correlated with depth, and is in agreement with the range in some well known
geothermal systems of the world (e.g. Bird and Spieler, 2004, Bird et al., 1988). Marks et al.
(2010), while studying epidote in the Reykjanes and Nesjavellir high-temperature
geothermal systems in Iceland observed an increase in the average molar ratio of pistachite
with depth. The authors interpreted this observation to indicate an increase in temperature.
In the Olkaria geothermal system, well OW-910 is a high-temperature well with regard to
measured down-hole formation temperatures, reaching a maximum of about 300oC
(Musonye, 2015). The distribution of formation temperature in this well, therefore, is
consistent with the relationship between the molar ratio of pistachite and temperature
advocated by Marks et al. (2010). Temperature exerts a major influence on the transition of
silicate phases, and consequently controls the phase equilibrium (Arnason et al., 1993). By
contrast, this correlation is contradictory with the results of epidote analyses from the Salton
Sea geothermal system by Bird et al. (1988). Based on their observations, the authors
established that epidote becomes more aluminous with increasing depth and temperature.
Both changes in fluid composition and temperature were attributed as the most influential
variables causing changes in epidote composition in that case.
Zoning of epidote mineral grains observed, for example by Marks et al. (2010) in the
Reykjanes and Nesjavellir geothermal systems in Iceland is also confirmed in the present
study (e.g. Figure 32). The explanation for compositional zoning patterns in epidote is an
intriguing problem. This is not suprising considering the sensitivity of Fe and Al contents of
epidote to slight changes in a variety of intensive and extensive thermodynamic variables
during hydrothermal activity. Zoning also depends directly on the initial ratio of the activities
of Fe3+ and Al3+ in the aqueous solution. In well OW-910, intragrain compositional zoning
seems to be dependent directly on local changes in fluid transport properties and intensive
thermodynamic variables (e.g. CO2 concentrations in fluid phase, pH, bulk rock composition
and ƒO2) during the thermal evolution of the geothermal system. Such changes result in the
formation of several spatially related phases which are not in chemical equilibrium. In this
respect, it is reasonable to assume that intrusion activity (e.g. basaltic and micro-granite)
observed in this well caused a surge in CO2 concentrations in the thermal fluids. This in turn
caused a change in the fluid composition by lowering the pH of the fluids, consequently
affecting aqueous speciation of Al3+ and Fe3+ complexes. While this is important in many
cases, compositional zoning patterns in some epidote mineral grains are caused by non-
equilibrium, irreversible processes (Arnason and Bird, 1992). In this case, the relative rates
84
of Al3+ and Fe3+ mass transfer accompanying the irreversible dissolution of common rock
forming minerals control epidote compositional zoning.
5.8.2 Chlorite
The Fe-rich nature of chlorites seems to be correlated to the relatively high Fe content of the
bulk rocks, as seen from the results of whole-rock analysis from wells OW-922 (Table 5)
and 910 (Musonye, 2015). This observation is however, not valid for the relatively Mg-rich
chlorites at 864 m depth from well OW-910. The most likely explanation for these
differences is the structural limitation in the assignment of Fe2+ into trioctahedral chlorite,
which favours the incorporation of Mg2+ (rather than Fe2+) under oxidizing conditions, as
demonstrated by Bryndzia and Scott (1987). If a Mg-rich solution, such as seawater-
dominated hydrothermal solution, had its way into a geothermal system, chlorite would be
expected to be Mg-rich. The relatively low quantities of Na, K and Ca probably rule out the
presence of corrensite layers in the analysed chlorite from Olkaria geothermal system.
As a result of the difficulty in obtaining precise chlorite analyses, occasioned by the
relatively porous and uneven surface, it was not feasible to draw any conclusions about the
relationship between chlorite composition and formation permeability. Results of the molar
ratio of Fe2+ / (Fe2+ + Mg) in chlorite presented in Table 9 are inconsistent with the
observation made by Lonker and others (1993) for wells RN-8 and 9 in the Reykjanes
geothermal system, Iceland. In their study, the authors reported well defined correlation
between chlorite composition and formation permeability. They observed a progressive
increase in Fe2+ / (Fe2+ + Mg) ratio and a decrease in Si contents within or in the vicinity of
aquifers in wells RN-8 and 9. Reverse correlations were however noted from the less
permeable formations. In well OW-922, 2236 m depth coincides with a small feed zone
(Table 4), while 1236 m depth in well OW-910 is slightly below a major feed zone
(Musonye, 2015).
85
6 Conclusion and Recommendations
6.1 Conclusion
The lithological units encountered in well OW-922 are consistent with the general sub-
surface stratigraphy of the Olkaria Volcanic Complex. Total absence of intrusive rocks and
a marked reduction in rhyolitic volcanism is, however, observed in this well. The two aspects
point towards the location of the well being outside the main Olkaria volcanic zone.
Generally, the well is dominated by low-temperature alteration minerals, which are observed
extending to greater depths. Abundant calcite deposition is apparent, especially towards the
bottom of the well, and the mineral is observed overprinting epidote.
Five small feed zones have been recognized in the well. They are connected to lithological
contacts, variations in alteration patterns, circulation losses, prevalence of calcite deposition,
intense pyritization, changes in ROP and kinks in temperature recovery logs.
Intense calcite deposition, failure of the well to accept water during injection test and the
absence of clear pivot point are indications that permeability in the well is very limited.
Hydrothermal alteration mineralogy shows that the area has had temperatures ≥ 240oC in the
past, as indicated by the presence of epidote. However, measured formation temperature
reached a maximum of 127oC after 49 days recovery period. Consequently, the difference in
measured formation and inferred alteration temperatures distinctly demonstrates that the
geothermal system around the well has significantly cooled over time. Calcite overprinting
epidote further argues in favour of this conclusion.
From petrochemical analysis, the rocks show a complete range of compositions from basalt
to trachyte or rhyolite. Highly evolved rocks are predominant.
Based on major oxides and trace elements systematics, fractional crystallisation is the
dominant magma differentiation mechanism for evolution of the rocks of the study well.
Other magma modification mechanisms likely played some role, but only to a limited extent.
Significant changes in the abundance of major oxides and trace elements have not taken
place during hydrothermal alteration processes. Equally, a comparison of surface and sub-
surface samples indicates the possibility of a petrogenetic connection caused by a common
differentiation process. The samples are, therefore, geochemically comparable.
The compositional variations of epidote from two different depth intervals in well OW-910
appear to be strongly influenced by the combined effects of temperature, ƒO2 and fluid
composition. Results of similar studies from other geothermal systems corroborates this fact.
The presence of zonation in epidote mineral grains indicates that fluid composition and
pressure or temperature conditions in the geothermal system have changed with time.
Due to limited number of epidote and chlorite samples, it was not possible to carry out a
stratigraphic correlation in the four wells based on major and minor element contents.
86
6.2 Recommendations
Additional temperature and pressure profiles need to be measured. As the well has been shut-
in for a period of over one and a half years, current temperature and pressure measurements
will provide a true reflection of the state of the system, as opposed to the 49 days recovery
profile used in this study.
Fluid inclusion studies need to be carried out and integrated with the inferred alteration and
measured (or calculated) formation temperatures. Several attempts were made to conduct
fluid inclusion analysis for the well but were unsuccessful due to problems with sample
preparation. The findings would serve as a better means of reconstructing the past history of
the geothermal system around well OW-922.
In-depth geological and geophysical studies of the area east of Domes field is still strongly
recommended if expansion of production drilling has to continue in the area. Some
geophysical studies have already been carried in the area but the findings are limited in my
view. The combination and comparison of the findings with other disciplines (i.e.
geochemistry) will be crucial in better understanding the system east of the Domes.
Well OW-922 should be listed amongst the reservoir monitoring wells in Olkaria. In
particular, monitoring of the drawdown changes in this well will be vital in establishing
whether or not there exists a connection between system around the well and the Olkaria
reservoir.
There is need to carry out radiogenic isotope studies (Pb-Sr-Nd-Hf) to gain valuable
information about the source of the rocks from the four wells, that is whether the rocks are
derived from similar or different parental magmas. In addition, radiogenic isotopes will
provide useful insights about crustal contamination during ascent of the magma.
87
References
Achauer, U. and Masson, F., 2002: Seismic tomography of continental rifts revisited: from
relative to absolute heterogeneities. Tectonophysics, 358, p. 17-37.
Allen, D.J., Darling, W.G., and Burgess, W.G., 1989: Geothermics and hydrogeology of the
southern part of the Kenya Rift Valley with emphasis on the Magadi-Nakuru area. British
Geological Survey, report SD/89/1, 68 pp.
Ambusso, W.J., and Ouma, P.A., 1991: Thermodynamic and permeability structure of
Olkaria Northeast field: Olkaria fault. Geothermal Resource Council Transactions, 15, p.
237-242.
Arnason, J.G., and Bird, D.K., 1992: Formation of zoned epidote in hydrothermal systems.
Water-Rock interaction, Stanford University, USA, p. 1473-1476.
Arnason, J.G., Liou, J. G, and Bird, D.K., 1993: Variables controlling epidote composition
in hydrothermal and low-pressure regional metamorphic rocks. Proceedings Abhand Geol.
Bund., Salzburg, Austria vol. 49, p. 17–25.
Baker, B.H., Mohr, P.A., and Williams, L. A. J., 1972: Geology of the Eastern Rift System
of Africa. Contrib. GSA Special Papers. Vol. 136, p. 1-68.
Baker, B.H. and Wohlenberg, J., 1971: Structural evolution of the Kenya Rift Valley. Nature
229, p. 538-542.
Bird, D. K. and Spieler, A. R., 2004: Epidote in Geothermal systems. Reviews in Mineralogy
and Geochemistry, Vol. 56, p. 235-300.
Bird, D.K., Cho, M., Janik, C.J., Liou, J.G., and Caruso, L.J., 1988: Compositional,
order/disorder, and stable isotope characteristics of Al-Fe epidote, State 2-14 Drill hole,
Salton Sea geothermal system. J. of geophysics. Res., 93, p. 13135-13144. Bosworth, W., 1987: Axis volcanism in the Gregory Rift, East Africa: implications for
models of continental rifting. Geology, 15: p. 397-400.
Bliss, C.M. 1979: Geology and petrochemistry of the Naivasha area, Kenya RiftValley.
Ph.D. thesis (unpublished), Lancaster University, p. 1-270.
Bragason, G, O., 2012: Strontium isotope shift in clay minerals, epidote and geothermal fluid
in the Hellisheiði Geothermal Field, SW-Iceland, University of Iceland, MSc. Thesis, p. 1-
45.
Browne, P.R.L., 1984: Lectures on geothermal geology and petrology. UNU-GTP, Iceland,
report 2, 92 pp.
Browne, P.R.L., 1978: Hydrothermal alteration in active geothermal fields. Annual Review
Earth and Planetary Sciences, 6, p. 229–250.
88
Bryndzia, L.T. and Scott, S.D., 1987: The composition of chlorite as a function of sulfur and
oxygen fugacity. An experimental study. Am. J. Sci., 287: p. 50-76.
Chorowicz, J., 2005: The East African Rift System. Journal of African Earth Sciences, 43,
p. 379-410.
Clarke, M.C.G., Woodhall, D.G., Allen, D., Darling, G., 1990: Geological, Volcanological
and hydrogeological controls of the occurrence of geothermal activity in the area
surrounding Lake Naivasha, Kenya. Min. of Energy, Nairobi, report 150, 138 pp. Davies, G.R., and Macdonald, R., 1987: Crustal influences in the petrogenesis of the
Naivasha basalt-comendite complex, combined trace elements and Sr-Pb isotope constraints.
Journal of petrology 28, p. 1009-1031.
Dubois, M., and Martinez-Serrano, R, G., 1998: Chemical variations in chlorite at the Los
Humeros geothermal system, Mexico. Clay and clay minerals, Volume 46, No. 6, p. 615-
628.
Ebinger, C., 2005: Continental breakup: the East African perspective. Astronomy and
Geophysics, 46, p. 216–221.
Ebinger, C. and Sleep, N. H., 1998: Cenozoic magmatism in central and east Africa resulting
from impact of one large plume. Nature 395, p. 788–791.
Franzson, H., Zierenberg, R. and Schiffman, P., 2008: Chemical transport in geothermal
systems in Iceland, Evidence from hydrothermal alteration. Journal of Volcanology and
Geothermal Research, Volume 173, p. 217-229.
Harðarson, B.S., 1993: Alkalic rocks in Iceland with special reference to the Snaefellsjokull
Volcanic System, Ph.D. thesis (unpublished), University of Edinburgh, p. 1-330.
Heumann, A., Davies, G.R., 2002: U-Th disequilibrium and Rb-Sr age constraints on the
magmatic evolution of peralkaline rhyolites from Kenya. Journal of Petrology 43, p. 557-
577.
KenGen 2014: Well completion report for well OW-922. KenGen in-house report
(unpublished), p. 1-32.
KenGen 2012: Stratigraphy and hydrothermal alteration mineralogy of well OW-915B.
KenGen internal report (unpublished), p. 1-40.
Kristmannsdóttir, H., 1979: Alteration of basaltic rocks by hydrothermal activity at 100-
300°C. In: Mortland, M.M., and Farmer, V.C. (editors), International Clay Conference1978.
Elsevier Scientific Publishing Co., Amsterdam, p. 359-367.
Laird, D.A., 2006: Influence of layer charge on swelling of smectites. Applied Clay Science,
34, p. 74-87.
89
Lagat, J.K., 2004: Geology, hydrothermal alteration and fluid inclusion studies of the Olkaria
Domes geothermal field, Kenya. University of Iceland, MSc. Thesis, UNU-GTP, Iceland,
report 2, 71 pp.
Le Bas, M.J., Le Maitre, R.W., Streckeisen, A., and Zanettin, B., 1986: A chemical
classification of volcanic rocks based on the total alkali-silica diagram: Journal of Petrology,
v. 27, no. 3, p. 745-750.
Lonker, S.W., Franzson, H., & Kristmannsdottir, H., 1993: Mineral-Fluid Interactions in
the Reykjanes and Svartsengi Geothermal Systems, Iceland. American Journal of Science,
293 (7), p. 605-670.
Macdonald, R., Black, H.E., Fitton, J.G., Marshall, A.S., Nejbert, K., Rodgers, N.W., and
Tindle, A.G., 2008: The roles of fractional crystallisation, magma mixing, crystal mush
remobilization and volatile-melt interactions in the genesis of a young basalt-peralkaline
rhyolite suite, the Greater Olkaria volcanic complex, Kenya Rift Valley. J. Petrology, 49-8,
p. 1515-1547.
Macdonald, R., Scaillet, B., 2006: The central Kenya peralkaline province: Insights into the
evolution of peralkaline salic magmas. J. Petrology, 2006, p. 59-73.
Macdonald, R., Rogers, N. W., Fitton, J. G., Black, S., and Smith, M., 2001: Plume-
Lithosphere interactions in the generation of basalts of the Kenya Rift, East Africa. Journal
of Petrology, vol. 42, p. 877-900.
Macdonald, R., 1974: Nomenclature and petrochemistry of the peralkaline oversaturated
extrusive rocks. Bull Volcanol 38: p. 498–516.
Mannvit/ÍSOR/Vatnaskil/Verkís Consortium, 2011: Consultancy services for geothermal
optimization study of the Greater Olkaria Geothermal Fields, Report 3: Revision of the
conceptual model of the Greater Olkaria Geothermal System – Phase I. Mannvit/ÍSOR/
Vatnaskil/Verkís, Reykjavík, November 2011, 100 pp.
Marks, N., Schiffman, P., Zierenberg, R.A., Franzson, H., and Fridleifsson, G.Ó., 2010:
Hydrothermal alteration in the Reykjanes geothermal system: Insights from Iceland deep
drilling program well RN-17. J. of Volcanology & Geothermal Research 189 (2010), p. 172–
190.
Marshall, A.S., Macdonald, R., Rogers, N.W., Fitton, J.G., Tindle, A.G., Nejbert, K., and
Hinton, R.W., 2009: Fractionation of peralkaline silicic magmas: the Greater Olkaria
volcanic complex, Kenya Rift Valley. J. Petrology, 50, p. 323-359.
Mielke, P., Nehler, M., Bignall, G., and Sass I., 2015: Thermo-physical rock properties and
the impact of advancing hydrothermal alteration- A case study from the Tauhura geothermal
field, New Zealand. Journal of Volcanology and Geothermal Research, vol. 301, p. 14-28.
Moore, M.D. and Reynolds, R. C. Jr., 1997: X-ray diffraction and the identification and
analysis of clay minerals. New York: Oxford University Press,18th ed., 378 pp.
90
Mortensen, A.K., Egilson, T., Gautason, B., Arnadottir S., and Gudmundsson A., 2014:
Stratigraphy, alteration mineralogy, permeability and temperature conditions of well IDDP-
1, Krafla, NE-Iceland. Geothermics, 49, p. 31-41.
Mungania, J., 1999: Geological report of well OW-714. Kenya Power Company internal
report.
Musonye, X.S., 2015: Sub-surface petrochemistry, stratigraphy and hydrothermal alteration
of the Domes area, Olkaria geothermal field, Kenya, University of Iceland, MSc. Thesis, p.
1-109.
Mwania, M.M., Munyiri, S., and Okech, E., 2013: Borehole geology and hydrothermal
mineralisation of well OW-35, Olkaria East geothermal field, Kenya, Report Number 1, p.
1-56.
Mwangi, D.W., 2012: Borehole geology and hydrothermal mineralisation of well OW-916,
Olkaria Domes geothermal field, Naivasha, Kenya. Reports 2012, Number 24, p. 541-571.
Nkapiani, M.N., 2014: Application of stable isotope geochemistry to tracing recharge and
flow systems of fluids in the Olkaria geothermal field, Kenya. Reports 2014, Number 24, p.
483-504.
Nyblade, A.A. and Robinson, S.W., 1994: The African superswell. Geophysical Research
Letters, 21, p. 765–768.
Ofwona, C., Omenda, P., Mariita, N., Wambugu, J., Mwawongo, G., and Kubo, B., 2006:
Surface geothermal exploration of Korosi and Chepchuk prospects. KenGen internal report,
44 pp.
Ojiambo, B.S., and Lyons, W. B., 1993: Stable isotope composition of Olkaria geothermal
field fluids, Kenya. Geothermal Resource Council, Transactions 17, p. 149-145. Okoo, J.A., 2013: Borehole geology and hydrothermal mineralisation of well OW–39A,
Olkaria East geothermal field, Naivasha, Kenya. Reports 2013, Number 24, p. 547-574.
Omenda, P.A., 1998: The geology and structural controls of the Olkaria geothermal system,
Kenya. Geothermics, 27-1, p. 55-74.
Omenda, P.A., 1997: The Geochemical Evolution of Quaternary Volcanism in the South-
Central Portion of the Kenya Rift: Ph.D. thesis, University of Texas at El Paso, p. 1-219.
Otieno, V., Okoo, J., Munyiri., S., Wanjohi, D., and Wanjohi A., 2014: Structural Mapping
of Olkaria South West Field, Olkaria Geothermal Project: KenGen Internal Report, 41 pp.
Otieno, V., and Kubai, R., 2013: Borehole geology and hydrothermal mineralisation of well
OW-37A, Olkaria east geothermal field, Kenya, Report Number 2, p. 57-105.
Pirajno, F., 1992: Hydrothermal Mineral Deposits. John Wiley and Sons, Sydney, 709 pp.
91
Peccerillo, A., Donati, C., Santo, A. P., Orlando, A., Yirgu, G., and Ayalew, D., 2007:
Petrogenesis of silicic peralkaline rocks in the Ethiopian rift: Geochemical evidence and
volcanological implications. Journal of African Earth Sciences 48, p. 161-173.
Reyes, A.G., 2000: Lectures on petrology and mineral alteration in hydrothermal systems:
From diagenesis to volcanic catastrophes. UNU-GTP, Iceland, report 18-1998, 77 pp.
Ring, U., 2014: The East African Rift System: Austrian Journal of Earth sciences, Vienna
2014. Vol. 107/1, p. 132-146.
Rollinson, H.R., 1995: Using geochemical data: Evaluation, presentation, interpretation.
Longman Group Ltd., England.
Ronga, F., Lustrini, M., Marzoli, A., and Melluso, L., 2009: Petrogenesis of a basalt-
comendite-pantellerite rock suite: the Boseti Volcanic Complex (Main Ethiopian Rift).
Miner Petrol, 98, p. 227-243.
Ronoh, I.J., 2012: Borehole geology and hydrothermal mineralisation of well OW-912B,
Olkaria geothermal field, Kenya. Report 29, p. 695-732.
Schiffman, P. and Roeske, S., 2014: Electron Microprobe Analysis of Minerals, University
of California, p. 293-306.
Shackleton, R.M., 1996: The final collision between East and West Gondwana: where is it?
Journal of African Earth Sciences, 23, p. 271-287.
Simiyu, M.S., 2010: Status of Geothermal Exploration in Kenya and Future Plans for its
Development. Proceedings World Geothermal Congress, 25-29 April 2010.
Simiyu, S.M., and Keller, G.R., 2000: Seismic monitoring of the Olkaria Geothermal area,
Kenya Rift valley, Journal of Volcanology and Geothermal Research 95, p. 197-208.
Simmons, S.F. and Christenson, B. W., 1994: Origins of calcite in boiling geothermal
systems. American Journal of Science, Vol. 294, p. 361-400.
Smith, M., Mosley, P., 1993: Crustal heterogeneity and basement influence on the
development of the Kenya Rift, East Africa. Tectonics 12 (2), p. 591–606.
Steiner, A., 1977: The Wairakei Geothermal Area, North Island, New Zealand: its sub-
surface geology and hydrothermal rock alteration. N. Z. Geol. Surv. Bull. 90, 127 pp.
Steiner, A., 1968: Clay minerals in hydrothermally altered rocks at Wairakei, New Zealand.
Clay and clay minerals, Vol. 16, p. 193-213.
Verma, S.P., Torres-Alvarado, I. S., Pandarinath, K., and Dulski, P., 2008: Element mobility
during the hydrothermal alteration of rhyolitic rocks of the Los Azufres geothermal field
(Mexico): Geothermic 37, p. 53-72.
92
Verma, S.P., Torres-Alvarado, I. S., Satir, M., and Dobson, P. F., 2005: Hydrothermal
alteration effects in geochemistry and Sr, Nd, Pb and O isotopes of magmas from the Los
Azufres geothermal field (Mexico): A statistical approach, Geochemical journal, Vol. 39, p.
141-163.
Weaver, B. D., Sceal, J.S.C and Gibson, I. L., 1972: Trace element data relevant to the origin
of trachytic and pantelleritic lavas in the East African Rift System. Contrib. Mineral. Petrol.
36, p. 181-194.
Williams, L.A.J., 1970: The volcanics of the Gregory Rift Valley, East Africa. Bulletin of
Volcanology, Vol. 34, p. 439-465. Wise, W.S., 2005: Zeolites. University of California-Santa Barbara, Santa Barbara, CA,
USA, Elsevier Ltd., p. 591-600.
Wilson, M., 1989: Igneous petrogenesis, A global tectonic approach. Department of Earth
Sciences, University of Leeds, p. 1-466.
Yoshio, W., Eiichiro, M. and Kozo, S., 2011: X-ray Diffraction Crystallography. New York:
Springer Heidelberg Dordrecht. p. 67-80.
93
Appendix A: Lithological description of well OW-922 based on binocular and
optical petrographic analyses.
0-46 m Pyroclastics: Light grey, unconsolidated heterogenous drill chips composed
predominantly of tuffaceous, pumiceous, obsidian, volcanic glass as well as obsidian. It is
moderately oxidized and no hydrothermal alteration is observed in this section. Alteration
minerals: oxides.
46-58 m: No samples (Loss of circulation returns).
58-96 m Pyroclastics: Light grey, considerably unconsolidated heterogenous drill cuttings
composed of tuffaceous, pumiceous, volcanic glass as well as obsidian. It is moderately
oxidized and no hydrothermal alteration is observed in this section. Alteration minerals:
oxides.
96-190 m: No samples (Loss of circulation returns).
190-302 m Pyroclastics: Light grey to yellowish-brown, unconsolidated heterogenous drill
chips composed predominantly of tuffaceous, pumiceous, volcanic glass as well as obsidian.
It is mildly oxidized. Alteration minerals: oxides, calcite.
302-350 m Tuffs (lithic): Brown to light grey, cryptocrystalline tuffaceous rock. It is highly
vesicular, vesicles are primarily infilled with zeolites (scolecite and mesolite) and rarely
calcite. Patches of fine grained green clays are also noted infilling vesicles. The rock is
weakly altered with minor pyrite disseminations in the matrix. Alteration minerals: clays,
calcite, zeolites.
350-362 m: No samples (Loss of circulation returns).
362-364 m Tuffs: Brown to light grey, cryptocrystalline tuffaceous rock. It is highly
vesicular, vesicles are primarily infilled with zeolites (scolecite and mesolite) and rarely
calcite. The rock is weakly altered with minor pyrite disseminations in the matrix. Alteration
minerals: clays, calcite, zeolites, pyrite.
364-408 m Trachyte: Grey, fine grained crystalline rock. It is phyric with sanidine and quartz
porphyries. Numerous specs of opaque minerals (olivine and pyroxenes) are noted as well
in the rock matrix. It depicts flow banding texture and slight intensity of alteration. Minor
pyrite is disseminated in the rock matrix. Alteration minerals: clays, zeolites, pyrite.
408-422 m Tuffs (lithic): Brown, highly vesicular tuffaceous rock. Vesicles are infilled with
chalcedony (412-414 m) and zeolites (scolecite and mesolite). The rock is weakly altered
with minor pyrite disseminations in the matrix. Alteration minerals: clays, calcite,
chalcedony, zeolites, pyrite.
422-476 m Trachyte: Grey, fine grained phyric rock with sanidine and quartz porphyries.
Specs of mafic minerals (olivine and pyroxenes) are noted in the rock matrix. It is slightly
94
vesicular, vesicles are infilled with calcite and chalcedony. Aveinlet infilled with chalcedony
noted between 446-448 m. Minor pyrite is disseminated in the rock matrix. Alteration
minerals: clays, calcite, titanite, chalcedony, pyrite.
476-498 m Tuff (lithic): Light grey to light brown, highly vesicular tuffaceous rock
intermixed with minor trachytic fragments. The latter are suspected to have fallen from the
unit above. Distinct vesicles infilled with secondary quartz (e.g. 476-478 m), zeolites (e.g.
480-482 m) and calcite are observed. Furthermore, wairakite is noted in a vesicle between
494-496 m.The rock is weakly altered with minor pyrite disseminations in the matrix.
Alteration minerals: clays, calcite, secondary quartz, wairakite, pyrite.
498-504 m Basalt: Dark grey, fine grained, slightly vesicular rock. Tuffaceous fragments
(suspected to have fallen from above) are notable in minor quantities. Quartz veins are
present between 500-502 m. It is slightly altered. Alteration minerals: clays, calcite,
secondary quartz.
504-564 m Trachyte: Grey, mixed cuttings of trachye (most abundant), tuffaceous and
basaltic fragments. It is veined and vesicular. Quartz vein noted between 516-518 m while
vesicles infilled with calcite are observed between 520-522 m. Alteration minerals: clays,
secondary quartz, calcite.
564-582 m Tuff (lithic): Mostly brown, whitish to brownish cryptocrystalline tuffaceous
fragments. It is mixed with pumiceous, trachytic and rhyolitic fragments and moderately
vesicular rock. Majority of the vesicles are empty, whereas, a few are infilled with zeolites
and secondary quartz. Pyrite is embedded as well as disseminated in the rock matrix. Minor
patches of green clays are noted replacing the matrix of individual fragments. Alteration
minerals: clays, calcite, pyrite, quartz, wairakite (574-578 m).
582-726 m Trachyte: Light grey, fine grained moderately porphyritic rock with feldspar and
quartz porphyries. Veinlets infilled with mafic minerals are discernible. Green clays are
observed replacing the mafic silicates. It is mildly oxidized and appears relatively fresh.
Partial loss between 596-604m and 636-638 m. Alteration minerals: clays, titanite.
726-744 m Tuff (lithic): Pale brown fine grained tuffaceous rock (dominant) mixed with
trachytic fragments. Occasional quartz phenocrysts are observed on both fragments. It is
moderately to highly oxidized but displays minimal intensity of alteration. Calcite is
observed infilling the vesicles. Alteration minerals: clays, calcite.
744-822 m Trachyte: Brown to dark grey, fine grained to cryptocrystalline rock. It is slightly
porphyritic with euhedral sanidine phenocrysts, some elongated in form. Mafic minerals are
outstanding in the matrix and show alteration to green clays. Albite noted replacing the
sanidine phenocrysts at 802-814 m and 816-818 m. It is mixed with minor tuffaceous
fragments. The formation shows minimal intensity of alteration. Alteration minerals: clays,
calcite, chalcedony, albite.
822-890 m Trachyte: Holocrystalline, grey to dark grey, fine grained rock depicting flow
banding texture (between 862-864 m). It is slightly porphyritic with subhedral to euhedral
phenocrysts of sanidine. Generally, the formation is massive, though vesicular between 870-
872 m, with green clays infilling the vesicles. Green clays are also observed extensively
replacing the mafic silicates. Albite is observed replacing sanidine phenocryst between 842-
844 m. The formation is slightly oxidized and exhibit slight degree of alteration. Alteration
minerals: clays, calcite, chalcedony, albite.
95
890-918 m Trachyte: Light grey to grey, fine grained poorly phyric rock with elongated
subhedral sanidine phenocrysts. It is relatively massive and holocrystalline. The rock is
slightly altered but shows high intensity of alteration between 904-910 m. Tuffaceous
fragments observed between 910-912 m (inferred to have fallen from above). Alteration
minerals: clays, calcite.
918-952 m Trachyte: Light grey, aphanitic rock. It is moderately feldspar phyric with the
dominant phenocryst phase being sanidine. Some phenocrysts are elongated and euhedral. It
is slightly altered. It is both slightly oxidized and altered. Alteration minerals: clays, calcite,
pyrite, albite.
952-966 m Tuff (lithic): Brown (952-962 m) to grey (962-966 m), fine grained tuffaceous
rock. It is highly vesicular between 962-966 m, vesicles are empty. It is moderately altered
to clays between 952-962 m and shows slight alteration for the remaining part. Alteration
minerals: clays, calcite, pyrite.
966-1046 m Trachyte: Grey, fine grained highly vesicular rock. It is moderately porphyritic
with subhedral sanidine phenocrysts. Most vesicles are empty and others partially infilled
with calcite. In addition, specs of mafic minerals are present, some being replaced by brown
and green clays. The formation is slightly altered but alteration intensity increases towards
the bottom (1024-1042 m). Partial loss between 1042-1046 m. Pyrite disseminated in the
rock matrix. Alteration minerals clays, pyrite, calcite, chalcedony (990-992 m).
1046-1150 m Rhyolite: Light grey to pink, fine grained massive and crystalline rock. It is
moderately porphyritic with quartz phenocrysts. Several specs of mafic minerals are noted
disseminated in the rock matrix. Some mafics are partially replaced by clays. The formation
is highly oxidized between 1054-1056 m and relatively fresh. Alteration minerals: clays,
pyrite.
1150-1198 m Basalt: Mostly brown, fine grained, mixed cuttings composed primarily of
basaltic fragments. It is slightly porphyritic with plagioclase phenocrysts. Majority of the
phenocryts have been replaced by clays whereas others are noted being replaced by calcite.
Mineral sequence noted between 1188-1190 m with chalcedony on the rim and calcite
precipitated on the core. Vesicles infilled with calcite are plentiful. Pyrite is disseminated in
the rock matrix. The formation is highly altered. Partial loss between 1152-1154 m and 1190-
1192 m. Alteration minerals: clays, pyrite, calcite, haematite (1192-1198 m).
1198-1214 m Basalt: Dark grey, fine grained to cryptocrystalline rock. It is poorly
porphyritic with indistinct plagioclase phenocrysts. Depicts slight intensity of alteration with
minor pyrite disseminations in the rock matrix. Alteration minerals: clays, pyrite, calcite,
haematite (1198-1200 m).
1214-1224 m Rhyolite: Light grey, fine grained highly felsic massive rock mixed with
cryptocrystalline basaltic fragments. It is phyric with quartz phenocrysts. Fresh specs of
mafic minerals are observed in the matrix. The formation is mostly fresh though slight
alteration to green and brown clays is evident. It is mildly oxidized as well. Alteration
minerals: clays.
1224-1260 m Trachyte: Light grey to pale green, fine grained holocrystalline rock but mostly
grey between 1234-1244 m. It is slightly porphyritic with sanidine phenocrysts. Minor
tuffaceous fragments noted between 1248-1260 m. The rock is slightly altered and
moderately oxidized between 1258-1260 m. Alteration minerals: titanite, clays, pyrite.
96
1260-1312 m Tuff (vitric): Mostly whitish, fine grained tuffaceous rock. It is highly vesicular
especially between 1294-1298 m, vesicles are infilled with calcite and green clays. Pyrite
dissemination (including euhedral cubes) is rather intense. This is a probable feed zone.
Alteration minerals: clays, calcite, pyrite.
1312-1440 m Trachyte: Light grey, to grey to pale green, fine grained rock. It is slightly
porphyritic with occasional subhedral sanidine phenocrysts. Calcite is noted replacing some
phenocrysts. Mafic silicates are being replaced by brown clays. Pyrite is noted infilling
veinlets as well as disseminated in the rock matrix. The formation is moderately altered.
Alteration minerals: pyrite, clays, titanite, calcite.
1440-1600 m Tuff (lithic): Grey to pale green, fine grained tuffaceous rock mixed with minor
trachytic fragments embedded within the lithic fragments. Vesicles infilled with calcite are
noted. It is slightly altered and pyrite disseminations are evident on the rock matrix. Partial
losses between 1454-1458 m,1462-1466 m and 1512-1518 m. Alteration minerals: pyrite,
clays, titanite, calcite.
1600-1642 m Tuff (lithic): Grey, fine grained tuffaceous rock. It is highly vesicular, vesicles
are infilled with clays and calcite. Also, veinlets are noted infilled with calcite. It shows
incipient alteration with minor pyrite disseminations in the rock matrix. Alteration minerals:
pyrite, clays, titanite, calcite.
1642-1654 m Basalt: Dark grey, fine grained crystalline rock. It is slightly porphyritic with
plagioclase phenocrysts. Occasionally, vesicles infilled with calcite are present. Also, calcite
is observed replacing some plagioclase phenocrysts. It is moderately altered and pyrite is
disseminated in the rock matrix. Alteration minerals: pyrite, clays, titanite, calcite.
1654-1700 m Basaltic trachyandesite: Brown, slightly porphyritic crystalline rock with
plagioclase phenocryts. The latter are noted being replaced by clays and calcite. Mafic
silicates display total alteration to clays. Platy calcite observed between 1646-1648 m and
1680-1682 m. Pyrite cubes are disseminated in the rock matrix. The formation is highly
altered. Alteration minerals: clays, calcite, pyrite, titanite.
1700-1730 m Trachydacite: Grey, fine grained to cryptocrystalline poorly porphyritic rock
with minor sanidine phenocrysts. Occasionally vesicles are present infilled with clays and
calcite. Epidote colouration is evident on a sanidine phenocryst between 1728-1730 m.
Pyrite is disseminated in the rock matrix. Partial circulation loss between 1722-1730 m. The
formation is moderately altered. Alteration minerals: clays, calcite, pyrite, titanite, epidote,
quartz.
1730-2052 m Trachyte: Grey to pale brown, fine grained slightly phyric rock with indistinct
sanidine phenocrysts. Vesicles infilled with calcite, green clays and secondary quartz are
noted. Veinlets infilled with calcite, mafic minerals and brown clays are noted as well.
Calcite replacing sanidine phenocryst between 1768-1770 m. Platy calcite noted between
1944-1946 m. Epidote coloration noted between 1704-1706 m, 1770-1774 m and 1930-1932
m. It is mildly oxidized and depicts moderate intensity of alteration. However, the formation
is highly altered (rated at a scale of 2.5) between 1788-1852 m. Minor pyrite disseminated
in the matrix. Partial losses between 1794-1796 m, 1798-1806 m, 1810-1816 m, 1820-1824
m, 1836-1838 m, 1842-1844 m, 1926-1930 m, 1954-1958 m, 1964-1972 m, and 2014-2018
m. A possible feed zone is suspected within this zone. Alteration minerals: pyrite, clays,
secondary quartz, titanite, calcite, chalcedony (2008-2010 m), epidote.
97
2052-2070 m Trachyte: Grey, fine grained to cryptocrystalline rock with minor sanidine
phenocrysts. It is relatively dense with specs of mafic minerals conspicous in the matrix,
some showing alteration to brown clays. Trace amounts of pyrite are disseminated in the
matrix. It is slightly altered. Alteration minerals: clays, calcite, pyrite.
2070-2186 m Trachyte: Grey to pale brown, fine grained slightly phyric rock with indistinct
sanidine phenocrysts. It is highly vesicular. Vesicles infilled with calcite and green clays are
noted. Veinlets are noted as well, infilled with calcite, mafic minerals and brown clays.
Chalcedony observed recrystallizing to secondary quartz between 2114 and 2116 m. It is
mildly oxidized and depicts moderate intensity of alteration. Partial loss between 2070-2076
m, 2104-2114 m, 2128-2130 m, 2142-2150 m, 2158-2162 m, 2170-2172 m, and 2184-2186
m. A possible feed zone within this zone. Alteration minerals: pyrite, clays, titanite, calcite.
2186-2210 m Trachydacite: Pale grey to pink, fine grained aphyric rock mixed with grey
trachytic fragments. The rock is rather massive and depicts slight intensity of alteration.
Minor pyrite is embedded in the rock matrix. Partial loss between 2186-2192 m and 2206-
2210 m. Alteration minerals: clays, calcite, pyrite.
2210-2326 m Trachyte: Mostly grey though showing distinctive gradation to brown between
2256-2260 m and 2302-2306 m. It is sparsely porphyritic with occassional sanidine
phenocrysts. Minor basaltic and tuffaceous fragments are noted. Vesicles and veinlets
infilled with brown clays are noted. Minor pyrite is disseminated in the rock matrix. Calcite
deposition is quite intense in this zone (scale 2.5), possibly reflecting a minor feed zone. It
is moderately to highly altered. Partial circulation returns experienced between 2214-2216
m, 2246-2248 m, 2266-2268 m, 2276-2278 m Alteration minerals: clays, haematite, pyrite,
calcite, titanite, epidote coloration (2234-2236 m).
2326-2370 m Trachydacite: Pink, highly felsic fine grained slightly porphyritic rock with
primarily quartz phenocrysts. Minor basaltic fragments are observed between 2334-2342 m.
Vesicles are noted infilled with secondary quartz and dark green clays. Veinlets are largely
infilled by brown clays and white minerals. The latter also replace the mafic silicates.
Euhedral pyrite cubes are disseminated in the matrix. It is moderately altered. Alteration
minerals: clays, pyrite, titanite, calcite, quartz.
2370-2592 m Rhyolite: Pale brown, fine grained, poorly porphyritic rock. It shows flow
texture and veinlets infilled with pyrite and brown minerals. Minor pyrite is disseminated in
the rock matrix. Generally, the formation is moderately to highly altered but shows slight to
moderate intensity of alteration from 2448 m to 2474 m. Partial loss of returns between 2374-
2380 m, 2388-2390 m, and 2398-2412 m. Alteration minerals: clays, pyrite, calcite,
haematite, chalcedony, epidote coloration (2570-2580 m).
2592-2780 m Trachydacite: Light grey to grey, fine grained slightly porphyritic rock with
subhedral to euhedral elongated sanidine phenocryts. It is moderately to strongly porphyritic
with striated sanidine phenocrysts between 2598-2610 m and 2618-2632 m. It is slightly
vesicular. Vesicles are infilled with brown clays and calcite. Pyrite is disseminated in the
rock matrix. The formation is moderately altered. Alteration minerals: clays, calcite, pyrite,
haematite,.
2780-2784 m Trachyandesite: Grey to dark grey, fine grained poorly porphyritic rock with
minor feldspar phenocrysts. It is relatively dense with numerous specs of mafic minerals
noted in the rock matrix. Epidote noted replacing feldspar phenocrysts (plagioclase) between
2782-2784 m. Paragenetic sequence, with epidote occuring in the core and calcite occuring
98
in the rim is also noted between the same depth interval. Minor pyrite disseminated in the
matrix. It shows slight degree of alteration. Alteration minerals: clays, calcite, pyrite, quartz,
epidote.
2784-2990 m Trachyte: Mostly grey, poorly porphyritic rock (though slightly porphyritic
between 2786-2794 m) with indistinct sanidine phenocrysts. It is slightly oxidized and both
vesicles and veinlets infilled with calcite and quartz are observed. Pyrite disseminations are
discernible in the matrix. Partial losses between 2816-2820 m. It is slightly altered.
Alteration minerals: clays, pyrite, haematite, titanite, calcite
99
Appendix B: XRD analysis for clay minerals in well OW-922.
OW-922-CLAY ANALYSIS RESULTS
Depth (m) d(001) A d(001) G d(001) H d(002) Mineral Type Other minerals
318-320 No clear peaks No clear peaks No clear peaks No clear peaks No clear peaks
338-340 14.11 19.13 9.89 Smectite Sm. Feldspars, Amphiboles
358-360 12.48 16.77 9.97 Smectite Sm. Feldspars, Amphiboles
380-382 15.36 17.89 10.01 Smectite Sm. Feldspars
398-400 13.85 17.27 9.91 Smectite Sm. Feldspars, Amphiboles
418-420 No clear peaks No clear peaks No clear peaks No clear peaks No clear peaks No clear peaks
438-440 15.66 18.05 10.11 Smectite Sm. Feldspars, Amphiboles
474-476 10.03 10.03 10.03 7.23, HIT=0 Illite + Chlorite (traces) Chl: ill Feldspars, Amphiboles
538-540 10.04 10.04 10.04 7.23, HIT=0 Illite + Chlorite (traces) Chl: ill Feldspars, Amphiboles
558-560 No clear peaks No clear peaks No clear peaks No clear peaks No clear peaks No clear peaks
738-740 9.96 9.96 9.96 Illite (traces) ill Feldspars, Amphiboles
758-760 10.03 10.03 10.03 Illite (traces) ill Feldspars, Amphiboles
778-780 10.14 10.14 10.14 Illite (traces) ill Feldspars, Amphiboles
952-960 No clear peaks No clear peaks No clear peaks No clear peaks No clear peaks No clear peaks
978-980 No clear peaks No clear peaks No clear peaks No clear peaks No clear peaks No clear peaks
998-1000 No clear peaks No clear peaks No clear peaks No clear peaks No clear peaks No clear peaks
1038-1040 15.41, 10.07 17.42, 10.07 10.07 7.18, HIT=0 Smectite+ Illite + Chlorite Sm:ill: Chl Feldspars, Amphiboles
1058-1060 13.69, 10.13 17.30, 10.13 10.13 7.31, HIT=0 Smectite + Illite Sm: ill Feldspars, Amphiboles
1198-1200 15.13 17.15 10.29 7.26, HIT=0 Smectite + Unst. Chlorite Sm: Chl Feldspars
1218-1220 12.92 17.24 9.86 7.07, HIT=0 Smectite + Unst. Chlorite Sm: Chl Feldspars
1340-1342 12.07 12.07 12.07 7.12, HIT=1 Unst. Chlorite + illite Chl: ill Feldspars
1358-1360 12.69, 10.06 17.09, 10.06 10.06 7.09, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars, Amphiboles
100
1418-1420 12.86, 9.96 17.09, 9.96 9.96 7.13, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars, Amphiboles
1438-1440 15.29, 9.99 17.17, 9.99 9.99 7.09, HIT=1 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars, Amphiboles
1458-1460 15.34, 9.79 17.08, 9.79 9.79 7.08, HIT=2 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars, Amphiboles
1554-1556 9.96 9.96 9.96 7.14, HIT=0 Chlorite + Illite Chl:ill Feldspars, Amphiboles
1558-1560 15.18, 10.11 17.26, 10.11 10.14 7.11, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars, Amphiboles
1576-1578 12.89, 9.91 17.14, 9.91 9.91 7.06, HIT=1 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars, Amphiboles
1578-1582 13.94, 9.93 16.99, 9.93 9.93 7.05, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars, Amphiboles
1598-1600 12.69, 10.03 17.08, 10.03 10.03 7.04, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars, Amphiboles
1618-1620 13.14, 9.84 16.99, 9.84 9.84 7.07, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars, Amphiboles
1640-1642 15.36 17.17 9.91 7.17, HIT=0 Smectite+ Unst. Chlorite Sm:Chl. Feldspars, Amphiboles
1658-1660 14.02, 10.01 17.11, 10.01 10.01 7.09, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars, Amphiboles
1696-1698 14.99, 9.98 16.9, 9.98 9.98 HIT=0 Smectite+Illite Sm:ill Feldspars
1698-1700 12.51, 10.19 17.17, 10.19 10.19 7.11, HIT=0 Smectite+ Illite+ Chlorite Sm:ill:Chl Feldspars
1716-1718 13.48, 10.07 17.57, 10.07 10.07 7.21, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars
1718-1720 13.01, 9.89 16.99, 9.89 9.89 7.16, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars
1738-1740 14.4, 9.98 16.99, 9.98 9.98 7.22, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars
1740-1742 13.5, 9.96 17.26, 9.96 9.96 7.13, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars
1760-1762 14.47, 10.11 17.26, 10.11 10.11 7.19, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars
1778-1780 12.96, 9.99 17.38, 9.99 9.99 7.1, HIT=1 Smectite+ Illite+Chlorite Sm:ill:Chl
1780-1782 13.32, 9.96 17.17, 9.96 9.96 7.07, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars
1800-1802 12.84, 9.91 16.54, 9.91 9.91 7.07, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars
1818-1820 12.64, 9.98 17.20, 9.98 9.98 7.09, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars
1838-1840 14.64, 9.89 16.63, 9.89 9.89 7.16, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl
1860-1862 15.23, 9.98 17.79, 9.98 9.98 7.15, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars
1876-1880 14.68, 10.05 17.17, 10.05 10.05 7.08, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars
1900-1902 14.19, 10.08 16.91, 10.08 10.08 7.08, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl
1918-1920 14.04, 9.92 14.04, 9.92 14.04, 9.92 7.06, HIT=0 Chlorite + Illite Chl: ill
1940-1942 14.53, 10.01 16.99, 10.01 10.01 7.13, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars
1956-1958 14.26, 9.98 17.63, 9.98 9.98 7.15, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars
1958-1960 9.98 9.98 9.98 7.15, HIT=0 Unst. Chlorite + Illite Chl: ill Feldspars
1978-1980 9.84 9.84 9.84 7.12, HIT=0 Unst. Chlorite + Illite Chl: ill Feldspars
1996-1998 10.03 10.03 10.03 7.09, HIT=0 Unst. Chlorite + Illite Chl: ill Feldspars
101
2018-2020 9.96 9.96 9.96 7.14, HIT=0 Unst. Chlorite + Illite Chl: ill Feldspars
2038-2040 13.11, 10.17 17.41, 10.17 10.17 7.19, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars
2056-2058 10.12 10.12 10.12 7.15, HIT=0 Unst. Chlorite + Illite Chl: ill Feldspars
2058-2060 13.16, 9.95 16.94, 9.95 9.95 7.05, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars
2078-2080 14.06, 9.97 17.14, 9.97 9.97 7.04, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl Feldspars
2096-2098 10.34 10.34 10.34 7.13, HIT=0 Unst. Chlorite + Illite Chl: ill Feldspars
2118-2120 14.28, 9.82 14.28, 9.82 14.28, 9.82 7.06, HIT=0 Chlorite + Illite Chl: ill Feldspars
2156-2158 9.94 9.94 9.94 7.13, HIT=0 Unst. Chlorite + Illite Chl: ill Feldspars
2196-2200 14.12, 9.96 17.12, 9.96 9.96 7.08, HIT=0 Smectite+ Illite+Chlorite Sm:ill:Chl
2198-2200 14.59, 10.09 14.59, 10.09 14.59, 10.09 7.13, HIT=0 Chlorite + Illite Chl: ill Feldspars
2240-2242 12.56 17.12 9.73 7.03, HIT=0 Smectite+ Chlorite Sm:Chl
2258-2262 14.73, 10.2 14.73, 10.2 14.73, 10.2 7.26, HIT=0 Chlorite + Illite Chl: ill Feldspars
2278-2282 15.2, 10.31 15.2, 10.31 15.2, 10.31 7.18, HIT=0 Chlorite + Illite Chl: ill Feldspars
2296-2300 14.02, 9.94 16.97, 9.94 9.94 7.07, HIT=0 Smectite+ + Illite+Chlorite Sm:ill:Chl
2316-2320 14.06, 10.16 16.94, 10.16 10.16 7.04, HIT=0 Smectite+ + Illite+ Chlorite Sm:ill:Chl
2338-2340 10.06 10.06 10.06 7.04, HIT=0 Unst. Chlorite + Illite Chl: ill
2338-2342 12.76, 9.89 16.94, 9.89 9.89 7.03, HIT=0 Smectite+ + Illite+Chlorite Sm:ill:Chl Feldspars
2354-2358 14.02, 10.11 17.26, 10.11 10.11 7.09, HIT=0 Smectite+ + Illite+ Chlorite Sm:ill:Chl
2458-2460 14.14, 10.14 17.39, 10.14 10.14 7.09, HIT=0 Smectite+ + Illite+ Chlorite Sm:ill:Chl
2482-2484 - - - 7.14, HIT=0 Unst. Chlorite Chl. Feldspars
2496-2500 10.01 10.01 10.01 7.07, HIT=0 Unst. Chlorite + Illite Chl: ill Feldspars
2516-2518 9.99 9.99 9.99 7.09, HIT=0 Unst. Chlorite + Illite Chl: ill Feldspars
2522-2524 10.86 10.86 10.86 7.12, HIT=0 Unst. Chlorite + Illite Chl: ill Feldspars
2536-2540 12.55, 9.99 17.03, 9.99 9.99 7.08, HIT=0 Smectite+ + Illite+ Chlorite Sm:ill:Chl Feldspars
2558-2562 14.45, 9.89 17.03, 9.89 9.89 7.09, HIT=0 Smectite+ + Illite+ Chlorite Sm:ill:Chl
2576-2580 13.01, 9.92 17.2, 9.92 9.92 7.11, HIT=0 Smectite+ + Illite+ Chlorite Sm:ill:Chl
2616-2618 12.48, 10.15 17.54, 10.15 10.15 7.08, HIT=0 Smectite+ + Illite+ Chlorite Sm:ill:Chl
2676-2678 14.77, 10.03 17.23, 10.03 10.03 7.08, HIT=0 Smectite+ + Illite+ Chlorite Sm:ill:Chl Feldspars
2696-2700 12.21, 9.91 17.14, 9.91 9.91 7.05, HIT=0 Smectite+ + Illite+ Chlorite Sm:ill:Chl
2716-2718 12.03, 9.91 12.02, 9.91 12.02, 9.91 7.09, HIT=0 Chlorite + Illite Chl: ill Feldspars
2758-2760 10.65 10.65 10.65 7.07, HIT=0 Unst. Chlorite + Illite Chl: ill Feldspars
2760-2762 10.72 10.72 10.72 7.09, HIT=0 Unst. Chlorite + Illite Chl: ill Feldspars
102
2798-2800 9.98 9.98 9.98 7.12, HIT=0 Unst. Chlorite + Illite Chl: ill Feldspars
2834-2836 14.36, 9.89 17.51, 9.89 9.89 7.13, HIT=0 Smectite+ + Illite+ Chlorite Sm:ill:Chl Feldspars
2840-2842 12.42, 10.06 17.15, 10.06 10.06 7.09, HIT=0 Smectite+ + Illite+ Chlorite Sm:ill:Chl Feldspars
2898-2900 13.16, 9.98 16.99, 9.98 9.98 7.04, HIT=0 Smectite+ + Illite+ Chlorite Sm:ill:Chl Feldspars
2918-2920 12.89, 10.14 17.54, 10.14 10.14 7.12, HIT=0 Smectite+ + Illite+ Chlorite Sm:ill:Chl Feldspars
2958-2960 10.05 10.05 10.05 7.07, HIT=0 Unst. Chlorite + Illite Chl: ill Feldspars
2978-2980 13.06, 9.95 17.26, 9.95 9.95 7.11, HIT=0 Smectite+ + Illite+ Chlorite Sm:ill:Chl Feldspars
103
Appendix C: ICP-OES analysis
procedure
The first step is to grind the drill chips in agate mortar to a fineness of approximately 100
mesh powder. Always remember to clean the mortar and pestle using aceton after
gringing each sample to eliminate contamination between the samples.
Weigh 100 mg ± 1mg of the sample in an epicure graphite crucible. Reset the scale to 0
g and add 200 mg ± 1 mg of flux (Lithium metaborate, LiBO2) to the sample. Mix the
two properly making sure that the mixture does not spill on the crucible walls. The
purpose of the flux is to increase the melting temperature of the sample.
Prepare the working standard samples. In this case, a total of six standard samples were
prepared. These include three external reproducibility standards (K-1919, BIR-1 and
BGM-1) and three internal calibration standards (A-THO, B-THO and B-ALK). 250 mg
± 1 mg of the sample and 500 mg ± 1 mg of the flux were used.
The samples were melted in an electric furnace heated at 1000oC for at least 30 mins and
thereafter left to cool for another 15-20 mins before the fused bead (or glass pellet) is
transferred into a bottle containing 30 ml of Rockan complexing acid. The latter, basically
is a mixture of de-ionized water, 5% HNO3, 1.33% HCL and 1.33% semi-saturated Oxalic
acid, H2C2O4. The three calibration standards were however, transferred into three
different bottles, each containing 75 ml of Rockan complexing acid. Highly acidic/highly
alkaline solutions are not recommended for this analysis because they will extinguish the
argon plasma.
The glass pellet and Rockan complexing acid are immediately run in a carousel to
complete solution to avoid silica precipitation. Complete solution is achieved between 2-
3 hours. The samples are now ready for analysis at the ICP-OES machine.
Analytical session began by running the three calibration standards. Spectra vision
software was applied in the calculation of two to three point calibration for every single
element. Instrumental reference sample (REF), used for monitoring drift during analysis
was made of equal parts (a third) of the internal calibration standards. The REF was
analysed at the beginning of the session and at 10 samples interval across the whole
analytical period. One calibration standard (B-THO) was analysed in between a batch of
10 samples to observe instrument reproducibility in multiple analyses.
During data processing, the raw data was copied and pasted into a correction spreadsheet.
Thereafter, all the batch analyses were normalized to 100%. The spreadsheet calculates
time dependent variation along each column of the analysis.
Application of the time-dependent variation-correction to the batch results to equal
absolute values of the reference samples, both at the beginning and the end.
105
Appendix D: EMP analysis for chlorite and epidote
1. QUANTITATIVE ANALYSIS RESULTS FOR CHLORITE
Representative EMP analyses of Chlorite-Well OW-917- 1870 m depth
Point 1 2 3 4 5 6 7 8 9 10
SiO2 22.78 22.62 22.26 19.58 24.59 22.93 22.67 22.62 22.67 20.09
TiO2 0.09 0.08 0.06 0.04 0.04 0.08 0.06 0.02 0.08 0.00
Al2O3 18.46 18.72 18.3 16.33 17.81 18.67 18.39 18.05 18.73 15.93
FeO 34.38 34.55 33.32 34.55 30.82 33.88 34.62 33.60 33.76 33.64
MnO 0.70 0.85 0.80 0.70 1.07 0.65 0.78 0.66 0.63 0.66
MgO 5.42 4.84 4.81 5.3 6.1 4.96 5.83 4.91 5.01 5.93
CaO 0.05 0.04 0.10 0.07 0.07 0.05 0.03 0.06 0.06 0.05
Na2O 0.42 0.33 0.44 0.47 1.07 0.42 0.39 0.39 0.29 0.33
K2O 0.58 0.39 0.56 0.58 0.65 0.61 0.47 0.53 0.50 0.62
SrO 0.00 0.00 0.02 0.00 0.00 0.04 0.00 0.00 0.12 0.06
Total 82.88 82.40 80.68 77.62 82.22 82.29 83.25 80.84 81.86 77.32
Number of cations on the basis of 24 Oxygens
Si 4.688 4.684 4.703 4.417 5.000 4.739 4.651 4.765 4.708 4.519
Ti 0.015 0.012 0.009 0.007 0.006 0.012 0.009 0.003 0.013 0.000
Al 4.480 4.569 4.558 4.342 4.267 4.548 4.447 4.481 4.586 4.225
Fe2+ 5.919 5.985 5.889 6.518 5.241 5.855 5.941 5.919 5.864 6.330
Mn 0.122 0.148 0.143 0.133 0.184 0.114 0.136 0.118 0.111 0.126
Mg 1.663 1.495 1.515 1.783 1.850 1.528 1.785 1.541 1.551 1.988
Ca 0.012 0.009 0.023 0.017 0.015 0.011 0.007 0.014 0.014 0.013
Na 0.166 0.131 0.182 0.207 0.424 0.168 0.156 0.161 0.119 0.146
K 0.152 0.103 0.150 0.167 0.170 0.160 0.122 0.142 0.133 0.178
Sr 0.000 0.000 0.003 0.000 0.000 0.005 0.000 0.000 0.015 0.008
Total 17.216 17.135 17.176 17.591 17.156 17.140 17.255 17.144 17.113 17.532
106
Representative EMP analyses of Chlorite-Well OW-922- 2236 m depth
Point 1 2 3 4 5 6 7 8 9 10
SiO2 22.20 20.04 20.23 21.01 18.88 19.86 21.38 21.52 22.54 19.71
TiO2 0.10 0.10 0.08 0.05 0.02 0.01 0.02 0.07 0.02 0.01
Al2O3 14.86 14.57 13.63 14.49 13.11 13.51 13.29 14.84 13.36 13.27
FeO 30.50 31.78 29.63 29.54 28.47 28.19 28.85 29.93 29.76 29.72
MnO 0.12 0.13 0.11 0.11 0.08 0.11 0.14 0.13 0.11 0.12
MgO 5.38 5.88 4.73 4.58 4.47 4.41 4.32 4.87 4.42 4.92
CaO 0.10 0.12 0.05 0.07 0.08 0.08 0.05 0.08 0.05 0.05
Na2O 0.75 0.51 0.32 0.32 0.30 0.32 0.27 0.33 0.33 0.29
K2O 0.32 0.31 0.31 0.32 0.38 0.45 0.29 0.48 0.27 0.34
SrO 0.03 0.00 0.00 0.00 0.00 0.05 0.00 0.00 0.00 0.00
Total 74.35 73.45 69.09 70.50 65.78 66.99 68.62 72.25 70.85 68.44
Number of cations on the basis of 24 Oxygens
Si 5.050 4.702 4.997 5.045 4.919 5.039 5.265 5.039 5.365 4.941
Ti 0.017 0.018 0.014 0.008 0.003 0.002 0.004 0.012 0.003 0.002
Al 3.985 4.031 3.968 4.102 4.026 4.041 3.857 4.096 3.748 3.920
Fe2+ 5.803 6.236 6.122 5.934 6.204 5.982 5.940 5.861 5.924 6.231
Mn 0.023 0.026 0.024 0.023 0.018 0.024 0.030 0.025 0.021 0.025
Mg 1.824 2.057 1.742 1.638 1.736 1.668 1.584 1.699 1.567 1.837
Ca 0.024 0.030 0.014 0.018 0.021 0.021 0.013 0.020 0.012 0.014
Na 0.329 0.233 0.151 0.151 0.151 0.159 0.127 0.152 0.153 0.143
K 0.092 0.094 0.098 0.099 0.125 0.146 0.093 0.144 0.082 0.109
Sr 0.004 0.000 0.000 0.000 0.000 0.007 0.000 0.000 0.000 0.000
Total 17.151 17.426 17.130 17.018 17.203 17.090 16.913 17.048 16.876 17.222
107
Representative EMP anayses of Chlorite-Well OW-922- 2782 m depth
Point 1 2 3 4 5 6 7 8 9
SiO2 23.42 32.83 23.41 22.12 20.96 24.07 30.84 27.84 20.75
TiO2 0.02 0.04 0.04 0.06 0.00 0.01 0.01 0.00 0.00
Al2O3 15.51 12.45 15.38 13.85 14.26 14.65 15.47 10.01 13.46
FeO 35.23 29.59 33.21 32.97 30.77 33.49 28.02 26.06 32.73
MnO 0.24 0.20 0.26 0.33 0.28 0.26 0.32 0.22 0.29
MgO 4.07 4.08 4.33 5.38 4.66 5.08 4.01 2.83 3.81
CaO 0.14 0.09 0.08 0.16 0.14 0.16 0.13 0.12 0.18
Na2O 0.09 0.14 0.33 0.18 0.17 0.08 1.69 0.17 0.12
K2O 0.13 0.11 0.12 0.13 0.18 0.14 0.12 0.19 0.12
SrO 0.07 0.00 0.00 0.00 0.00 0.03 0.00 0.00 0.08
Total 78.92 79.54 77.17 75.18 71.41 77.97 80.61 67.44 71.55
Number of cations on the basis of 24 Oxygens
Si 5.092 6.618 5.154 5.050 5.007 5.241 6.139 6.681 5.029
Ti 0.003 0.007 0.007 0.011 0.000 0.002 0.002 0.001 0.000
Al 3.975 2.960 3.991 3.727 4.015 3.760 3.629 2.832 3.844
Fe2+ 6.405 4.989 6.116 6.296 6.148 6.098 4.664 5.232 6.633
Mn 0.044 0.035 0.049 0.064 0.057 0.047 0.055 0.044 0.060
Mg 1.318 1.226 1.421 1.831 1.658 1.650 1.189 1.013 1.376
Ca 0.034 0.019 0.018 0.039 0.035 0.038 0.027 0.031 0.046
Na 0.038 0.055 0.143 0.080 0.078 0.035 0.651 0.078 0.058
K 0.035 0.029 0.033 0.039 0.053 0.038 0.030 0.059 0.039
Sr 0.009 0.000 0.000 0.000 0.000 0.004 0.000 0.000 0.012
Total 16.953 15.937 16.932 17.136 17.051 16.913 16.386 15.971 17.097
108
Representative EMP analyses of Chlorite-Well OW-910- 864 m depth
Point 1 2 3 4 5 6 7 8 9 10
SiO2 24.59 25.16 23.22 24.92 24.25 25.25 25.02 25.04 25.90 23.93
TiO2 0.03 0.02 0.04 0.03 0.05 0.00 0.03 0.00 0.01 0.04
Al2O3 13.23 13.51 12.57 12.86 11.94 13.14 12.95 13.68 13.34 12.34
FeO 12.16 12.24 12.67 12.46 12.02 12.20 12.60 11.96 12.33 12.03
MnO 0.21 0.26 0.20 0.19 0.18 0.21 0.19 0.23 0.19 0.21
MgO 17.05 16.63 17.30 16.72 16.21 16.70 17.58 15.63 15.73 17.10
CaO 0.29 0.28 0.27 0.37 0.32 0.31 0.31 0.29 0.30 0.33
Na2O 0.00 0.06 0.04 0.02 0.03 0.01 0.03 0.01 0.02 0.04
K2O 0.02 0.00 0.02 0.01 0.02 0.01 0.02 0.01 0.02 0.01
SrO 0.06 0.08 0.00 0.02 0.00 0.03 0.00 0.00 0.00 0.01
Total 67.64 68.24 66.33 67.60 65.02 67.85 68.73 66.84 67.83 66.04
Number of cations on the basis of 24 Oxygens
Si 5.428 5.496 5.276 5.509 5.573 5.542 5.446 5.559 5.668 5.425
Ti 0.005 0.003 0.007 0.006 0.009 0.000 0.006 0.000 0.002 0.006
Al 3.441 3.477 3.367 3.350 3.233 3.399 3.323 3.582 3.440 3.297
Fe2+ 2.245 2.235 2.408 2.304 2.310 2.238 2.294 2.222 2.257 2.281
Mn 0.039 0.048 0.038 0.035 0.035 0.039 0.035 0.042 0.035 0.040
Mg 5.609 5.415 5.861 5.511 5.553 5.462 5.703 5.174 5.133 5.777
Ca 0.069 0.067 0.065 0.088 0.078 0.072 0.073 0.068 0.070 0.081
Na 0.000 0.024 0.019 0.009 0.014 0.004 0.013 0.003 0.007 0.019
K 0.004 0.000 0.005 0.002 0.006 0.002 0.005 0.002 0.005 0.003
Sr 0.008 0.010 0.000 0.003 0.000 0.004 0.000 0.000 0.000 0.002
Total 16.85 16.77 17.05 16.82 16.81 16.76 16.90 16.65 16.62 16.93
109
Representative EMP analyses of Chlorite-Well OW-910- 1236 m depth
Point 1 2 3 4 5 6 7 8 9 10
SiO2 30.45 31.82 35.11 33.18 30.43 29.62 37.91 33.01 30.81 32.81
TiO2 0.11 0.03 0.06 0.03 0.00 0.03 0.00 0.02 0.02 0.01
Al2O3 16.90 19.12 14.33 17.43 16.52 16.50 12.86 16.46 18.40 17.96
FeO 17.62 17.25 16.94 17.42 17.50 18.37 17.56 16.25 17.61 18.29
MnO 0.28 0.26 0.25 0.25 0.27 0.27 0.29 0.25 0.31 0.20
MgO 15.59 18.07 16.05 14.52 18.64 17.54 15.99 16.23 19.82 16.19
CaO 0.44 0.39 0.94 0.85 0.56 0.40 3.37 1.45 0.32 0.44
Na2O 0.09 0.25 0.06 0.20 0.16 0.05 0.04 0.39 0.03 0.16
K2O 0.01 0.04 0.04 0.22 0.06 0.02 0.05 0.10 0.03 0.17
SrO 0.02 0.02 0.00 0.05 0.04 0.00 0.01 0.01 0.02 0.00
Total 81.50 87.25 83.78 84.15 84.18 82.80 88.08 84.16 87.35 86.23
Number of cations on the basis of 24 Oxygens
Si 5.634 5.463 6.239 5.902 5.469 5.439 6.452 5.863 5.316 5.717
Ti 0.015 0.004 0.008 0.004 0.000 0.004 0.000 0.002 0.002 0.002
Al 3.686 3.870 3.003 3.655 3.500 3.572 2.581 3.445 3.743 3.689
Fe2+ 2.726 2.477 2.519 2.592 2.631 2.822 2.499 2.413 2.542 2.666
Mn 0.043 0.038 0.038 0.037 0.041 0.042 0.042 0.038 0.045 0.029
Mg 4.298 4.626 4.252 3.85 4.993 4.803 4.058 4.297 5.096 4.206
Ca 0.087 0.072 0.179 0.162 0.109 0.078 0.614 0.276 0.058 0.082
Na 0.031 0.082 0.020 0.068 0.057 0.017 0.013 0.133 0.009 0.054
K 0.004 0.009 0.009 0.050 0.014 0.005 0.012 0.023 0.007 0.037
Sr 0.002 0.002 0.000 0.005 0.004 0.000 0.001 0.001 0.002 0.000
Total 16.53 16.64 16.27 16.33 16.82 16.78 16.27 16.49 16.82 16.48
110
Representative EMP analyses of Chlorite-Well OW-905A- 1560 m depth
Point 1 2 3 4 5 6 7 8 9 10
SiO2 25.68 25.61 24.72 26.12 25.19 24.12 24.99 26.53 24.96 25.54
TiO2 0.03 0.00 0.05 0.01 0.00 0.04 0.05 0.00 0.08 0.02
Al2O3 19.48 19.16 19.08 19.53 18.94 22.21 18.95 19.79 18.92 19.29
FeO 35.60 34.41 35.64 35.47 36.21 31.47 36.55 35.91 35.22 35.78
MnO 2.20 2.02 2.20 2.13 2.26 1.98 2.36 2.25 2.38 2.25
MgO 7.08 7.87 7.23 7.44 7.08 7.87 7.55 7.36 8.25 7.84
CaO 0.18 0.11 0.13 0.13 0.11 0.13 0.12 0.15 0.15 0.10
Na2O 0.09 0.02 0.03 0.08 0.07 0.21 0.03 0.09 0.02 0.00
K2O 0.02 0.02 0.02 0.03 0.01 0.04 0.02 0.03 0.03 0.03
SrO 0.02 0.02 0.00 0.00 0.07 0.06 0.00 0.08 0.00 0.00
Total 90.39 89.26 89.09 90.94 89.94 88.13 90.62 92.19 90.00 90.85
Number of cations on the basis of 24 Oxygens
Si 4.802 4.821 4.714 4.837 4.766 4.533 4.703 4.851 4.701 4.757
Ti 0.004 0.000 0.007 0.001 0.000 0.006 0.007 0.000 0.011 0.002
Al 4.294 4.250 4.288 4.264 4.224 4.920 4.203 4.264 4.200 4.235
Fe2+ 5.568 5.418 5.684 5.495 5.729 4.946 5.752 5.490 5.546 5.574
Mn 0.348 0.322 0.355 0.335 0.362 0.315 0.376 0.349 0.379 0.355
Mg 1.973 2.209 2.054 2.054 1.998 2.206 2.117 2.006 2.315 2.177
Ca 0.037 0.023 0.027 0.026 0.022 0.026 0.025 0.030 0.029 0.021
Na 0.032 0.009 0.010 0.028 0.027 0.077 0.012 0.031 0.008 0.000
K 0.005 0.005 0.004 0.008 0.002 0.009 0.004 0.006 0.006 0.008
Sr 0.002 0.003 0.000 0.000 0.008 0.007 0.000 0.009 0.000 0.000
Total 17.07 17.06 17.14 17.05 17.14 17.05 17.20 17.04 17.20 17.13
111
2. QUANTITATIVE ANALYSIS RESULTS FOR EPIDOTE
Representative EMP analyses of Epidote-Well OW-922- 2782 m depth
Point 1 2 3 4 5 6 7 8 9 10
SiO2 37.48 37.50 37.34 37.65 37.71 37.56 37.58 37.23 37.30 37.60
TiO2 0.00 0.03 0.00 0.07 0.01 0.00 0.03 0.04 0.09 0.04
Al2O3 24.62 24.71 24.60 24.40 24.71 24.43 24.48 24.43 24.59 24.29
Fe2O3 12.71 12.48 12.49 12.44 12.59 12.57 12.38 12.46 12.29 12.44
MnO 0.11 0.17 0.13 0.13 0.19 0.14 0.14 0.16 0.09 0.17
MgO 0.11 0.05 0.08 0.08 0.07 0.07 0.07 0.08 0.10 0.07
CaO 22.97 22.81 22.91 22.94 22.90 22.48 22.65 22.73 22.64 22.50
Na2O 0.02 0.03 0.00 0.04 0.06 0.03 0.03 0.04 0.06 0.07
K2O 0.04 0.04 0.05 0.04 0.05 0.05 0.04 0.03 0.05 0.06
SrO 0.36 0.32 0.51 0.44 0.41 0.46 0.50 0.46 0.27 0.55
Total 98.43 98.13 98.10 98.23 98.71 97.79 97.91 97.66 97.49 97.79
Number of cations on the basis of 24 Oxygens
Si 5.694 5.707 5.695 5.729 5.710 5.736 5.733 5.702 5.708 5.747
Ti 0.000 0.003 0.000 0.008 0.001 0.000 0.004 0.004 0.010 0.005
Al 4.409 4.432 4.422 4.377 4.411 4.399 4.402 4.410 4.436 4.376
Fe 3+ 1.454 1.429 1.433 1.423 1.435 1.444 1.422 1.436 1.416 1.430
Mn 0.015 0.021 0.016 0.017 0.024 0.018 0.018 0.021 0.012 0.023
Mg 0.024 0.012 0.018 0.018 0.016 0.016 0.016 0.017 0.024 0.016
Ca 3.740 3.720 3.743 3.740 3.716 3.679 3.703 3.730 3.713 3.685
Na 0.006 0.010 0.000 0.013 0.018 0.010 0.010 0.013 0.017 0.020
K 0.008 0.007 0.009 0.008 0.010 0.010 0.009 0.006 0.009 0.011
Sr 0.032 0.028 0.045 0.039 0.036 0.040 0.044 0.041 0.024 0.049
Total 15.382 15.370 15.380 15.372 15.378 15.352 15.360 15.380 15.369 15.361
112
Representative EMP analyses of Epidote-Well OW-910- 864 m depth
Point 1 2 3 4 5 6 7 8 9 10
SiO2 38.00 37.67 37.70 37.80 37.74 37.92 37.32 38.14 37.49 37.98
TiO2 0.05 0.09 0.01 0.01 0.06 0.11 0.24 0.02 0.11 0.11
Al2O3 24.18 22.99 23.45 23.92 23.93 24.54 22.21 25.57 20.46 25.29
Fe2O3 13.37 15.02 14.38 13.90 13.68 13.28 16.14 11.86 18.56 12.01
MnO 0.02 0.09 0.00 0.06 0.10 0.08 0.14 0.04 0.09 0.06
MgO 0.13 0.13 0.06 0.08 0.06 0.06 0.09 0.06 0.11 0.06
CaO 23.23 23.16 23.38 23.33 23.62 23.38 23.21 23.15 22.80 23.38
Na2O 0.00 0.06 0.00 0.01 0.00 0.01 0.00 0.00 0.00 0.02
K2O 0.01 0.00 0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.01
SrO 0.19 0.00 0.16 0.09 0.21 0.27 0.10 0.22 0.00 0.22
Total 99.17 99.22 99.14 99.21 99.41 99.64 99.46 99.08 99.63 99.14
Number of cations on the basis of 24 Oxygens
Si 5.731 5.714 5.714 5.712 5.699 5.697 5.681 5.721 5.733 5.707
Ti 0.005 0.011 0.001 0.001 0.007 0.012 0.028 0.002 0.013 0.013
Al 4.299 4.109 4.190 4.260 4.259 4.345 3.984 4.521 3.688 4.478
Fe 3+ 1.517 1.714 1.640 1.581 1.554 1.501 1.849 1.339 2.136 1.358
Mn 0.002 0.012 0.000 0.008 0.012 0.010 0.018 0.005 0.012 0.008
Mg 0.030 0.030 0.014 0.019 0.014 0.013 0.021 0.014 0.025 0.013
Ca 3.753 3.764 3.797 3.778 3.822 3.764 3.785 3.721 3.736 3.763
Na 0.000 0.018 0.000 0.003 0.000 0.002 0.000 0.001 0.000 0.005
K 0.002 0.000 0.000 0.000 0.000 0.000 0.001 0.002 0.000 0.002
Sr 0.016 0.000 0.014 0.008 0.019 0.023 0.009 0.020 0.000 0.019
Total 15.355 15.372 15.370 15.370 15.386 15.368 15.375 15.346 15.343 15.366
113
Representative EMP analyses of Epidote-Well OW-910- 1236 m depth
Point 1 2 3 4 5 6 7 8 9 10
SiO2 37.64 35.24 36.30 39.07 46.13 44.57 36.16 35.77 35.48 35.81
TiO2 0.58 0.57 0.67 0.44 0.49 0.56 0.47 0.49 0.39 0.34
Al2O3 8.73 7.67 7.75 9.15 9.18 6.66 8.03 7.23 6.92 7.19
Fe2O3 18.66 20.21 20.37 19.53 14.88 21.70 19.91 21.11 21.34 22.18
MnO 0.49 0.51 0.44 0.28 0.26 0.42 0.48 0.39 0.36 0.30
MgO 0.50 0.62 0.41 0.40 0.60 0.36 0.35 0.29 0.32 0.49
CaO 30.05 32.50 32.36 29.69 22.75 24.41 32.59 33.05 33.16 32.44
Na2O 0.88 0.03 0.21 1.07 0.17 0.06 0.02 0.01 0.04 0.06
K2O 0.17 0.07 0.07 0.07 3.59 0.62 0.03 0.05 0.09 0.04
SrO 0.02 0.04 0.00 0.05 0.00 0.01 0.00 0.01 0.10 0.05
Total 97.71 97.47 98.58 99.74 98.05 99.37 98.04 98.41 98.19 98.90
Number of cations on the basis of 24 Oxygens
Si 6.146 5.873 5.958 6.219 7.191 6.957 5.961 5.916 5.900 5.897
Ti 0.071 0.072 0.083 0.053 0.058 0.066 0.058 0.061 0.048 0.042
Al 1.681 1.506 1.499 1.716 1.686 1.225 1.559 1.409 1.356 1.395
Fe 3+ 2.293 2.536 2.517 2.339 1.745 2.549 2.471 2.628 2.670 2.750
Mn 0.068 0.072 0.061 0.038 0.034 0.055 0.067 0.055 0.051 0.041
Mg 0.121 0.155 0.099 0.095 0.140 0.084 0.086 0.070 0.078 0.120
Ca 5.258 5.804 5.692 5.064 3.801 4.082 5.756 5.857 5.909 5.725
Na 0.279 0.010 0.067 0.329 0.051 0.018 0.005 0.004 0.013 0.020
K 0.035 0.015 0.015 0.013 0.713 0.123 0.006 0.010 0.019 0.008
Sr 0.002 0.004 0.000 0.004 0.000 0.001 0.000 0.001 0.009 0.005
Total 15.953 16.047 15.992 15.871 15.418 15.160 15.970 16.012 16.054 16.003
114
Representative EMP analyses of Epidote-Well OW-905A- 1560 m depth
Point 1 2 3 4 5 6 7 8 9 10
SiO2 37.12 36.39 36.95 38.03 37.29 37.88 34.16 37.32 38.01 37.42
TiO2 0.14 0.08 0.23 0.11 0.08 0.11 0.34 0.13 0.16 0.11
Al2O3 23.10 23.48 19.55 24.18 24.02 24.34 20.39 24.09 24.44 24.18
Fe2O3 14.86 13.78 19.53 13.82 13.30 13.73 15.45 13.41 13.75 13.15
MnO 1.53 1.05 0.94 0.93 1.03 1.01 0.62 1.46 1.61 0.69
MgO 0.05 0.02 0.02 0.02 0.01 0.03 0.03 0.03 0.01 0.00
CaO 21.50 22.16 21.76 22.39 21.79 21.90 21.88 21.51 21.69 22.42
Na2O 0.00 0.01 0.01 0.02 0.01 0.00 0.00 0.00 0.00 0.00
K2O 0.01 0.01 0.00 0.01 0.03 0.03 0.05 0.03 0.01 0.02
SrO 0.05 0.01 0.06 0.04 0.00 0.03 0.09 0.00 0.02 0.09
Total 98.35 97.00 99.05 99.54 97.55 99.07 93.01 97.98 99.71 98.08
Number of cations on the basis of 24 Oxygens
Si 5.693 5.647 5.720 5.725 5.721 5.722 5.599 5.709 5.715 5.711
Ti 0.016 0.010 0.027 0.012 0.009 0.012 0.042 0.015 0.019 0.012
Al 4.175 4.295 3.566 4.291 4.343 4.334 3.940 4.342 4.331 4.349
Fe 3+ 1.715 1.609 2.275 1.566 1.536 1.561 1.906 1.544 1.555 1.510
Mn 0.199 0.139 0.124 0.118 0.133 0.129 0.086 0.189 0.205 0.090
Mg 0.011 0.005 0.004 0.004 0.003 0.007 0.008 0.007 0.003 0.000
Ca 3.534 3.684 3.610 3.612 3.581 3.546 3.842 3.525 3.495 3.666
Na 0.000 0.002 0.002 0.005 0.002 0.001 0.000 0.000 0.000 0.000
K 0.001 0.002 0.000 0.002 0.005 0.006 0.010 0.007 0.001 0.004
Sr 0.004 0.001 0.005 0.003 0.000 0.003 0.009 0.000 0.002 0.008
Total 15.348 15.393 15.333 15.338 15.334 15.321 15.442 15.337 15.326 15.350